Effects of DDT on Amyloid Precursor Protein Levels and Amyloid Beta Pathology: Mechanistic Links to Alzheimer's Disease Risk.

BACKGROUND: The interaction of aging-related, genetic, and environmental factors is thought to contribute to the etiology of late-onset, sporadic Alzheimer's disease (AD). We previously reported that serum levels of p,p'-dichlorodiphenyldichloroethylene (DDE), a long-lasting metabolite of the organochlorine pesticide dichlorodiphenyltrichloroethane (DDT), were significantly elevated in patients with AD and associated with the risk of AD diagnosis. However, the mechanism by which DDT may contribute to AD pathogenesis is unknown.
OBJECTIVES: This study sought to assess effects of DDT exposure on the amyloid pathway in multiple in vitro and in vivo models.
METHODS: Cultured cells (SH-SY5Y and primary neurons), transgenic flies overexpressing amyloid beta (Aβ), and C57BL/6J and 3xTG-AD mice were treated with DDT to assess impacts on the amyloid pathway. Real time quantitative polymerase chain reaction, multiplex assay, western immunoblotting and immunohistochemical methods were used to assess the effects of DDT on amyloid precursor protein (APP) and other contributors to amyloid processing and deposition.
RESULTS: Exposure to DDT revealed significantly higher APP mRNA and protein levels in immortalized and primary neurons, as well as in wild-type and AD-models. This was accompanied by higher levels of secreted Aβ in SH-SY5Y cells, an effect abolished by the sodium channel antagonist tetrodotoxin. Transgenic flies and 3xTG-AD mice had more Aβ pathology following DDT exposure. Furthermore, loss of the synaptic markers synaptophysin and PSD95 were observed in the cortex of the brains of 3xTG-AD mice. DISCUSSION: Sporadic Alzheimer's disease risk involves contributions from genetic and environmental factors. Here, we used multiple model systems, including primary neurons, transgenic flies, and mice to demonstrate the effects of DDT on APP and its pathological product Aβ. These data, combined with our previous epidemiological findings, provide a mechanistic framework by which DDT exposure may contribute to increased risk of AD by impacting the amyloid pathway. https://doi.org/10.1289/EHP10576.

Introduction

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder, which currently affects individuals.[1,2] Patients with AD accumulate extracellular amyloid beta () plaques and neurofibrillary tangles in the brain, which form the basis for the pathological diagnosis of AD. Plaques comprise fragments of various lengths (i.e., and ) formed by the sequential cleavage of the amyloid precursor protein (APP) by the enzymes and .[3]

AD can be categorized into two main types, early and late onset. Early-onset AD typically occurs at of age and involves a mutation in one of three genes: APP, presenilin-1 (PSEN1), or presenilin-2 (PSEN2).[4] The late-onset form (occurring at of age) is sporadic in nature and afflicts women to a greater extent than men, and the risk of development increases with age.[5] Late-onset AD has no clear genetic etiology, with the exception of increased risk in those carrying the apolipoprotein allele (). The risk associated with carrying the allele is dose dependent, with each allele increasing the risk of developing AD, as well as lowering the age of onset.[6]

Recently, attention has been drawn to the potential role for environmental factors as contributors to AD, though research in this area remains in its infancy.[4,7] Although some studies have argued for the role of occupational exposures to metals and solvents in the etiology of AD, the epidemiological evidence for these factors is not well supported by the current literature.[4,8] Rather, the most convincing epidemiological evidence supports a potential role for air pollution (particulate matter)[8-10] and pesticide exposure.[11-13] Numerous epidemiological studies have identified associations between occupational exposure to pesticides and lower neuropsychological,[14] behavioral,[15] and cognitive scores.[16] Unfortunately, most of these studies are limited, given that they do not specify individual pesticides and often rely on self-reported exposure.[13,17]

Dichlorodiphenyltrichloroethane (DDT) was one of the most widely used pesticides in the United States between the 1940s and 1970s.[12] DDT has been banned since 1972 in the United States, although it has been occasionally used in response to recurrent disease outbreaks.[18] Its long-lived metabolite, dichlorodiphenyldichloroethylene (DDE), is still present in human serum samples, as reported by the Centers for Disease Control and Prevention’s National Health and Nutrition Examination Survey (NHANES) as of 2016.[19] The presence of DDT/DDE in serum is not surprising because DDT/DDE has a long half-life in the environment and bioaccumulates in the food chain.[20] Exposure to DDT still occurs in many regions of the world, such as South Africa, where indoor residual spraying (IRS) is used to control malaria.[21] Indeed, recent reports from the Venda Health Examination of Mothers, Babies and their Environment (VHEMBE) birth cohort indicate five to seven times higher serum DDE levels in mothers who lived in a home with previous IRS compared with those living in homes that were not sprayed.[22] In addition, DDT is commonly used in countries (e.g., India) for agricultural purposes.[23] Its persistence in the environment and current use in different parts of the world continue to make DDT a potential health risk for individuals. In the United States, continued exposure likely comes from food imported from other countries where DDT is still used or from legacy contamination in the United States (e.g., soil and waterways).[24] Our group has reported that serum levels of p,p′-DDE were higher in patients with AD compared with age-matched control participants.[11,12] Further, an -fold increase in AD risk and worsened mini-mental state exam scores were modified by apolipoprotein E (APOE) genotype, with showing greater risk.[12] Similar findings have also been reported in populations in Delhi[25] and regions of north India,[25,26] as well as in Manitoba, Canada.[17]

Collectively, these data point to a potential role of DDT as a risk factor for AD. However, the mechanism by which DDT may contribute to this risk is unknown. In our previous study, we found that neuroblastoma cells exposed to DDT and DDE had higher APP protein levels.[12] Here, we sought to investigate potential mechanisms by which exposure to DDT may increase AD-related pathologies, focusing on the amyloid pathway using several in vitro and in vivo approaches. Taken together, these data provide a mechanistic framework responsible for the association between DDT exposure and increased AD risk.

Materials and Methods

Animals

Animal studies were reviewed and approved by the animal care and use committees at Florida International University, Northeast Ohio Medical University, and Rutgers Robert Wood Johnson Medical School. All studies complied with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines and were carried out in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.[27] C57BL/6J and 3xTG-AD, male and female, mice were purchased from the Jackson Laboratory and bred in-house. Three to four mice were housed per cage in a normal 12-h light/dark cycle, at and in relative humidity. For the 3xTG-AD mice, both males and females were used to assess brain levels of . Because of the sex-specific differences observed, all further experiments were carried out in 3xTG-AD female mice. Mice had access to food (standard chow) and water ad libitum.

Transgenic Drosophila melanogaster Fly Strains and Rearing

Drosophila melanogaster strains and , which overexpress , were generated in M.K.’s laboratory according to previously established protocols.[28] Briefly, the elavGal; strain was obtained by crossing virgin female elavGal (catalog no. 458) flies to male () flies. This strain overexpresses in the central nervous system (CNS) under the control of the yeast-derived transcriptional activator UAS/Gal4.[29] The () strain allows for the expression of in the photoreceptor cells of the eyes, which is driven ectopically under the control of the glass multiple receptor (GMR) promoter.[28] The Hikone-R (HR) strain (catalog no. 4267) that does not express was used as the control strain. elavGal and HR fly strains were purchased from the Bloomington Drosophila Stock Center (http://flystocks.bio.indiana.edu) and cultured on conventional sugar, cornmeal, and yeast medium at 25°C or 29°C. Male flies were used for all experiments in the present study owing to limitations in the availability of female flies.

Reagents

All chemical reagents were purchased from Sigma-Aldrich, with the exception of p,p′-DDT (Chem Service; CN-10876; 99.20% pure) and tetrodotoxin (TTX; Alomone Labs; catalog no. T-500). All cell culture reagents were purchased from Invitrogen.

Immortalized Cell Culture and Differentiation

SH-SY5Y human neuroblastoma cells were cultured in complete growth media consisting of Dulbecco’s Modified Eagle Medium supplemented with 10% heat-inactivated fetal bovine serum (FBS), 50 IU penicillin, streptomycin and l-glutamine. For experimentation, cells were plated in poly-d-lysine–coated 6-well plates at cells per well. After 24 h, the culture medium was replaced with complete growth medium containing 1% FBS and all-trans retinoic acid. Cells were differentiated for 6 d, and the culture medium was changed every 3 d.

Primary Neuronal Culture and Isolation

Brains were extracted from postnatal day (PND) 0–1 mixed-sex C57BL/6J pups, and the hippocampi and cortices were separately dissected and collected, as described previously.[30] Cells were dissociated with papain and dispase, then washed, triturated, pelleted, and resuspended in culture medium, after which they were seeded onto poly-d-lysine–coated 12-well plates at 500,000 cells/well for real-time quantitative polymerase chain reaction (RT-qPCR) analysis, and 45,000 cells/well in poly-l-lysine/laminin–coated black-walled 96-well plates for immunocytochemistry (ICC). Cells were maintained in neurobasal medium supplemented with B-27 plus, 50 IU penicillin, streptomycin and l-glutamine. Twenty-four hours after isolation, half of the media was changed to remove cellular debris. At 4 d in vitro, half of the media was replaced with culture media supplemented with 2% B-27 plus supplement and cytosine (Ara-C; Sigma; catalog no. C1768), which inhibits mitotic cells from proliferating.[31] Half of the culture media was replaced every 3 d.

In Vitro Exposure

Both SH-SY5Y and primary hippocampal and cortical neurons were treated with p,p′-DDT and TTX at the indicated concentrations. p,p′-DDT stock was prepared in dimethyl sulfoxide (DMSO) and TTX stock in pure water. For SH-SY5Y cells, treatments were diluted in complete growth medium containing 1% FBS for 24 or 72 h. For primary neurons, treatments were diluted in complete neurobasal medium and treated for 24 h. Control cells were treated with 0.1% DMSO in treatment medium, depending on cell type, for the same durations. For treatments including TTX, cells were pretreated for 15 min with TTX, followed by the addition of the respective treatments.

Cytotoxicity Assay

Cell viability was evaluated with SYTOX Green Nucleic Acid Stain (Invitrogen; catalog no. S7020). SH-SY5Y cells were plated in 96-well plates at cells/well, as aforementioned, and treated with p,p′-DDT (0.1, 0.5, 1, or ) or 0.1% DMSO (vehicle control) for 72 h. Primary hippocampal and cortical neurons were plated at 5,000 cells/well in a 96-well plate and treated at the previously mentioned concentrations for 72 h. For the SYTOX Green assay, following treatment, cells were incubated in SYTOX for 15 min at room temperature protected from light. Cytotoxicity was measured using fluorescence excitation at and emission at on the SpectraMax microplate reader (Molecular Devices). passages were used for SH-SY5Y, and isolations for primary neurons.

Measurement of in Media from SH-SY5Y Cells

Following 24 h treatment, culture media was collected and secretion of into media was measured using a commercial enzyme-linked immunosorbent assay (ELISA) kit as per the manufacturer’s instructions (Invitrogen; catalog no. KHB3481). The absorbance at was determined using a SpectraMax microplate reader (Molecular Devices). experiments.

Fly Survival Analysis

Life span and survival of the male and flies were assessed as previously described, specifically to determine whether DDT was lethal in each strain.[32] Briefly, DDT was dissolved in ethanol and administered with fly food. Media enriched with only 0.01% ethanol was used as a control. Twenty to 30 flies were placed in a food vial. Each vial was kept on its side at 25°C, in 70% humidity, and under a 12-h light-dark cycle. Flies were transferred to fresh media (control or DDT-supplemented) every 3 d. The number of flies alive was assessed every 3 d. experiments, each experiment consisting of flies.

Transgenic Flies Eye Phenotype Analysis

displays a rough eye phenotype due to the accumulation of in the photoreceptors of the eyes in a concentration-dependent manner.[28] The HR strain that does not express was used as the control strain. Male flies were collected at 0–3 d of age and raised for 30 d at 29°C on DDT in 0.01% ethanol. Control flies were raised on a diet with 0.01% ethanol in standard fly media. Flies were transferred to fresh media every 3 d. The eye phenotype was recorded at wk 0, wk 2, and wk 4 by temporarily immobilizing the flies on ice and visualizing the eyes under a light stereomicroscope at magnification (Leica Stereomicroscope). The rough eye phenotype exhibits variable penetrance, and flies were assigned to three different categories according to the extent to which the surface area of the eye displayed clear evidence of ommatidia disorganization. The phenotype was categorized as mild, moderate, or severe. To be assigned to these categories, fly eyes needed to exhibit disorganization in a third of the eye (mild), half of the eye (moderate), or the entire eye (severe). experiments, each experiment consisted of 100 individuals.

In Vivo Exposure in Wild-Type Mice

C57BL/6J male and female mice at 14 wk of age were obtained from the Jackson Laboratory. Mice were housed 4 per cage and administered p,p′-DDT in corn oil by oral gavage every 3 d for 30 d. Control mice were administered corn oil by oral gavage on the same schedule. This protocol causes no overt toxicity, is four times lower than the lowest median lethal dose () reported in mice[33] and produced brain levels of DDT similar to those observed in the adipose tissue of the U.S. population in the late 1960s.[34,35] Mice were housed in a normal 12-h light/dark cycle with access to food and water ad libitum. Animals were sacrificed by isoflurane inhalation and saline perfusion, followed by decapitation 1 d after the last dose of DDT. A subset of the saline-perfused brains was fixed in 10% formalin for 7 d and cryoprotected in a 30% sucrose solution diluted with phosphate-buffered saline (PBS) and 0.1% sodium azide. Another subset of brains was removed, and the hippocampus and frontal cortex were dissected on ice. These samples were immediately frozen in liquid nitrogen for subsequent protein and RNA isolation. Animal handling and experiments were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals[27] and approved by the animal care committee of Florida International University.

In Vivo Exposure in Triple-Transgenic (3xTG-AD) Mice

Male and female 3xTG-AD mice, from 9 to 12 months of age, a time frame that incorporates the earliest stages of amyloid pathology,[36,37] were administered p,p′-DDT in corn oil by oral gavage ( gavage volume) every 3 d for 90 d. Control animals were administered corn oil by oral gavage at the same frequency and duration. Animals were sacrificed at 12 months of age by isoflurane administration, followed by decapitation. The left hemisphere of the brain was fixed in 10% formalin and cryoprotected, and the hippocampus and cortex were dissected and frozen in liquid nitrogen for subsequent protein analysis. All procedures were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals.[27] and approved by the Institutional Animal Care and Use Committee of the Northeast Ohio Medical University.

RT-qPCR

RNA was isolated using TRIzol reagent (Thermo Fisher; catalog no. 15596026), following the manufacturer’s instructions. Complementary deoxyribonucleic acid (cDNA) was synthesized with RNA and iScript cDNA Synthesis Kit for tissue (Bio-Rad; catalog no. 1708891) and SuperScript II for cells (Thermo Fisher; catalog no. 18064014). Primer sets were checked with the National Center for Biotechnology Information Basic Local Alignment Search Tool and ordered from Thermo Fisher. Subsequent PCR reactions were performed using Itaq Universal SYBR green (Bio-Rad; catalog no. 1708891), following the manufacturer’s instructions. RT-qPCR was performed in duplicate on the Quantstudio 6 Flex System (Applied Biosystems). The following program parameters were used to amplify the target gene–polymerase activation/initial denaturation at 95°C for 10 min, followed by denaturation at 95°C for 15 s and annealing/extension at 60°C for 1 min for 40 cycles. Dissociation and melt curves were run to ensure that single amplicon peaks were obtained without nonspecific amplification. Samples were normalized as previously described,[38] except TATA-box binding protein or expression was monitored in parallel samples as housekeeping genes. Data were calculated and analyzed using the method.[39] For primer sequences, see Table S1. passages for SH-SY5Y cells, isolations for primary neurons, and animals/group for C57BL/6J brain tissue samples.

Immunocytochemistry

ICC was performed as previously described.[40] Briefly, cells were permeabilized with 0.1% Triton X-100, blocked with 2% bovine serum albumin (BSA) for 1 h and incubated overnight in a cocktail of APP (Y188; Abcam; catalog no. 32136; 1:1,000) and microtubule-associated protein 2 antibodies (Abcam; catalog no. 5392; 1:1,000). Cells were then incubated with the appropriate Alexa Fluor secondary antibodies; goat antirabbit 488 at 1:1,000 (Thermo Fisher; catalog no. A-11034) and goat antichicken 594 at 1:1,000 (Thermo Fisher; catalog no. A-11042) for 1 h at room temperature. For SH-SY5Y cells, images were taken using a Carl Zeiss Axiophot El- Einsatz microscope with a ProgResVR C14plus camera and ProgResVR CapturePro 2.8 software using a lens [numerical aperture (N.A.) 0.50]. Optical density per intensity of fluorescence against APP stain was semiquantified in individual cells using Image-Pro Plus 7.0 software (Media Cybernetics, Inc.). The values represent from 50–60 cells/well from three individual experiments, each performed in triplicate, and calculated as the percentage of control. For the primary hippocampal and cortical neurons, images were taken on Keyence BZ-X810 microscope using a lens (N.A. 0.60). Optical density per intensity of fluorescence of APP stain was semiquantified in individual cells using the BZ-X800 Analyzer Hybrid Cell Count Module (Keyence). Parameters were set to a threshold of control and applied automatically to all images within the proper comparisons. Integrated density was calculated using brightness and cell counts by the software. The values represent from 60–90 cells/well from eight images taken per well, in duplicate from four individual experiments and calculated as the percentage of control. Specificity was determined by omission of primary or secondary antibodies.

Immunohistochemistry

Immunohistochemistry (IHC) of the Drosophila brain was performed to stain for levels. Flies were raised for 21 d at 29°C on fly media supplemented with DDT; control flies were raised with 0.01% ethanol. Flies were anesthetized by carbon dioxide inhalation. Whole brains were dissected from the elavGal; flies at 3 wk of age under a stereomicroscope by carefully opening the head region using fine-tipped forceps. Fly brains were collected in 0.5% Tween-20 in PBS (PBST) and then fixed for 30 min at room temperature in 4% paraformaldehyde. The brains were incubated in 70% formic acid for 10 min to expose the epitope and then blocked for 1 h with 5% normal goat serum. Brains were incubated overnight at 4°C with 6E10 (Covance; catalog no. SIG-39300) primary monoclonal antibody at 1:3,000 dilution, followed by an Alexa Fluor 488 goat antimouse antibody at 1:130 for 2 h at room temperature. Images were taken using a Zeiss microscope. Analysis was performed using ImageJ software.[41] Images were threshold to control, and integrated density was measured. Each brain area was then measured using the draw tool and integrated densities were then normalized to respective brain areas. brains/group.

Coronal sections were cut from the left hemisphere of the brain at (3xTG-AD) on a freezing sliding microtome (Thermo Fisher) and stored in freezer storage solution (30% ethylene glycol, 30% glycerol, 30% distilled water in PBS) at [42] Every eighth section was pulled for 3xTG-AD throughout the hippocampus. Sections were washed with PBS and then blocked with 5% normal goat serum for 1 h at room temperature. 3xTG-AD sections were incubated with 1:1,000 of (Invitrogen; catalog no. 700254) antibody overnight at 4°C. Sections were washed with PBS, and they were incubated in a biotinylated secondary antibody at 1:200 dilution (Vector Laboratories; catalog no. BA-1000) for 1 h at room temperature, then rinsed, followed by 1 h incubation in the avidin–biotin peroxidase complex (Vector Laboratories; catalog no. PK-4000). After another rinse, sections were incubated with 3,3′-diaminobenzidine (DAB; Vector Laboratories; catalog no. SK-4100) with nickel enhancer. Images were taken using the Leica DM2500 LED microscope using a lens (N.A. 0.25) and oil lens (N.A. 1.4). Plaque size was quantified in 3xTG-AD mice in sections stained for by DAB staining. Briefly, two to three sections were imaged per animal for the cortex, CA1 and subiculum with a lens (N.A. 0.45). Images were then analyzed using the BZ-X800 Analyzer Hybrid Cell Count Module with permutations set for area (in micrometers squared). Parameters were set to the threshold of control and applied automatically to all images, then images were analyzed with the software and plaque area was calculated per plaque. animals/group for 3xTg-AD.

APP was visualized by immunofluorescent staining in C57BL/6J male and female mice exposed to DDT every 3 d for 30 d. Briefly, thick sections containing the frontal cortex and hippocampus were washed with PBS and blocked with 2% BSA, 0.1% Triton X-100, 0.01% PBST for 1 h at room temperature. Sections were then incubated in rabbit anti-APP (Y188, 1:1,000 in blocking solution) overnight at 4°C. The following day, sections were washed with PBS and stained with Alexa Fluor goat antirabbit (1:1,000) for 1 h at room temperature. Images were taken at magnification using the BZ-X800 microscope (Keyence).

APOE expression was evaluated in the 3xTG-AD brains by immunofluorescent staining. Four sections from both the frontal cortex and hippocampus region were tested in two to three brains per group. The sections were permeabilized with 0.5% Triton X-100 in PBS for 20 min, incubated in citrate buffer (Sigma-Aldrich; catalog no. c9999; pH 6.0) at 80°C for 20 min for antigen retrieval, and blocked in PBS with 10% BSA and 0.1% Triton X-100. Sections were then incubated overnight at 4°C with 1:500 dilution of rabbit monoclonal APOE (Abcam; catalog no. 183596) and 1:500 dilution of chicken polyclonal glial fibrillary acidic protein (GFAP) antibody (Invitrogen; catalog no. PA11004) to colocalize the APOE expression in the astrocytes. The following day, sections were incubated with 1:1,000 dilution of Alexa Fluor 488 and 594 conjugated secondary antibodies (Thermo Fisher; catalog no. A-11034 and catalog no. A-11042) for 1 h at room temperature. Trueblack (Biotium; catalog no. 23007) was applied before mounting the sections to reduce background autofluorescence. Images were taken at magnification using the BZ-X800 microscope (Keyence). The APOE expression was quantified by using ImageJ software. The same minimum and maximum threshold values were applied to all images and the integrated density of APOE was measured. Statistical significance () was determined by unpaired Student’s -test. Statistics were not run for the cortex because there were only two samples available for the DDT mice. For the full table of antibodies and dilutions, see Table S2.

Western Immunoblotting

Protein was extracted and quantified as previously described.[43] Briefly, protein was extracted using Tris lysis buffer [ sodium chloride (NaCl), Tris, ethylenediaminetetraacetic acid (EDTA), ethylene glycol-bis(-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 1% Triton X-100; pH 7.5], with 2% protease (Pierce; catalog no. 78439) and phosphatase inhibitors (Sigma-Aldrich; catalog no. P-5726 and catalog no. P0044). For APP, of total protein for the cortical samples and of total protein for the hippocampal samples were used from C57BL/6J male and female mice. For PSD95, of cortical and hippocampal samples were used, whereas for synaptophysin, cortical and hippocampal samples were used from 3xTG-AD female mice. All samples were separated on 4–12% Tris-Bis NuPage Mini gels (Fisher Scientific; catalog no. NP0323BOX). Proteins were then transferred to polyvinylidene fluoride (PVDF) membranes using the iBlot2 system (Invitrogen) at 20V for 4 min. Membranes were then blocked for 1 h in 5% BSA solution and then incubated overnight using the following antibodies: APP (Abcam; catalog no. 32136, 1:1,000)/PSD95 (Cell Signaling Technology; catalog no. 3450; 1:5,000)/Synaptophysin (Abcam; catalog no. 32127; 1:20,000) in 2.5% BSA. Membranes were then washed with Tris buffered saline with Tween 20 (TBST) and incubated with the appropriate secondary antibody (1:10,000 for APP and synaptophysin, 1:5,000 for PSD95) for 1 h at room temperature. This was detected using chemiluminescence and visualized using the Bio-Rad Molecular Imager ChemiDoc Imaging System. Membranes were then stripped with Thermo Fisher Scientific Restore PLUS Western Blot Stripping Buffer for 15 min (Fisher Scientific; catalog no. PI46430), blocked, and reprobed with (Sigma; catalog no. A5441; 1:10,000) for 1 h at room temperature to ensure equal loading of samples. The densitometric values were calculated using the ImageJ software and normalized to values prior to statistical analyses. Total protein was measured using Pierce Bicinchoninic Acid (BCA) Protein Assay Kit (Thermo Fisher; catalog no. 23227). animals/group for C57BL/6J males and females and 3xTG-AD females.

Quantification of Hippocampal Volume

Quantitative analyses for control and DDT-treated ( every 3 d) 3xTG-AD female mice (12 months old, saline: , DDT: ) were performed using computer-assisted stereology with a Leica DM 2500 LED microscope equipped with a digital camera and StereoInvestigator software (version 11; MBF Bioscience) by a single, blinded observer. Initial subsampling techniques were performed to determine appropriate sampling parameters.[44] Volumes (in micrometers cubed) for the CA1, CA2, CA3, and dentate gyrus (DG) subfields, as well as for the entire hippocampus, were measured using the area fraction fractionator (AFF) probe, founded on a Cavalieri point-counting system. Using a objective, markers were placed on a grid of points () overlaying the sampling area. Every point on the grid received two markers: one for the hippocampus and one of four potential subregion markers for the CA1, CA2, CA3, and DG. Volume was calculated as , and percentage (%) of hippocampal volume occupied by each subregion was determined from estimated area fractions calculated by the AFF probe. Volumetric measurements were taken across the rostral–caudal axis of the hippocampus from a minimum of 10 brain sections per mouse. The mean number of sampling sites for each area per animal was .

Meso Scale Discovery Multiplex Assay

Hippocampal and cortical samples were sonicated in 200 or Tris lysis buffer ( NaCl, Tris; pH 7.5, EDTA, EGTA, 1% Triton X-100), with 2% protease and phosphatase inhibitors. The human Amyloid Peptide Neurodegeneration Panel 1 [meso scale discovery (MSD); 6E10; catalog no. K15200G] assay was used to obtain measurements. The samples were run in duplicate with of total protein loaded per well. The samples were incubated overnight on a shaker at 4°C, and the remainder of the procedure was performed without modification. The assay was read using the MSD QuickPlex SQ 120, and data were normalized to total protein. animals/group.

Measurement of Oligomeric

Because of the significant pathology observed in 3xTG-AD female mice, but not in male mice, we quantified oligomeric from female samples. Frozen cortices and hippocampi from 3xTG-AD control and DDT-exposed female mice were homogenized in Tris lysis buffer, and total protein was quantified. Oligomeric () was quantified per the manufacturer’s protocol by a sandwich ELISA that uses the MOAB-2 antibody (Biosensis; catalog no. BEK-2215-2P). This antibody specifically detects oligomeric peptides and not APP. Briefly, of samples and standards were added to the capture antibody–coated wells. Oligomerized provided by the manufacturer was used as a positive control and the plate was incubated overnight at 4°C. The next day, wells were washed and the detection antibody was added for 1 h at room temperature. This was followed by the addition of streptavidin–horseradish peroxidase conjugate for 30 min at room temperature after which 3,3′,5,5′-tetramethylbenzidine (TMB) substrate was added to generate a detectable blue signal. After 10 min, stop solution was added and the absorbance was read at 450 nm. Data were analyzed using a four parametric logistic curve fit to assess in samples and further normalized to total protein concentration. Comparisons between control and DDT-exposed mice were analyzed by unpaired Student’s -test, and data are represented as . animals/group.

Statistical Analysis

Statistical analyses were performed using GraphPad Prism (version 7.05; GraphPad Software). Data were assessed for normality by the D’Agostino–Pearson normality test and the Shapiro–Wilk normality test. Unpaired Student’s -test was used when analyzing data where two means were compared with only one variable. One-way and two-way analyses of variance (ANOVA) were used when appropriate (multiple treatment doses) or (two or more variables), with Bonferroni post hoc tests. Differences were considered statistically significant at . Data are expressed as . For the in vitro SH-SY5Y and primary culture experiments, refers to each independently performed experiment. For experiments conducted in the Drosophila strains, alterations in life span length due to accumulation were evaluated using the Kaplan–Meier curve. Variation in Kaplan–Meier curves were assessed for statistical significance using a log-rank survival test, as described by Bland and Altman.[45] Eye degeneration was recorded through three trials as previously detailed, wherein each trial ommatidia degeneration was assayed by two independent observations. Mean values for eye phenotype progression were plotted with SD to assess significance in the difference between DDT-exposed and control flies. Significance was assayed through a chi-square test. For data involving animal studies, refers to the number of individual animals used in the study and is indicated in the respective figure legends. When possible, samples were blinded to the investigator or identified strictly using a code and unblinded following data analysis, especially in the case of in vivo methods, including the hippocampal volume measurements, IHC staining, as well as MSD assay results.

Results

APP Expression and Secretion in SH-SY5Y Cells Exposed to DDT

To determine the effects of DDT on cellular markers of AD, SH-SY5Y cell cultures were treated with DDT for 72 h and expression was examined. DDT concentrations were selected based on the range of serum concentrations found in exposed human populations during the 1960s in the United States and in currently exposed populations in countries with active DDT spraying for mosquito eradication.[20] Treatment of SH-SY5Y cells with DDT did not induce cell death, as determined by SYTOX assay; however, cells exposed to DDT exhibited -fold higher cell death compared with controls (Figure 1A; Table S3). Therefore, further experiments were conducted with DDT. At 24 h, SH-SY5Y cells treated with the highest concentration of DDT () had significantly higher mRNA levels of APP (, ), BACE1 (, ), PSEN1 (, , a gene for a subunit of involved in regulating its activity[3]), neprilysin (NEP; , ), and APOE (, ) (Figure 1B; Table S4).

Figure 1.

Effect of DDT treatments on (A) cell viability, (B) expression of AD-related genes, (C,D) APP protein levels, and (E) secreted in SH-SY5Y cells. (A) Sytox assay (percentage cell death) data after 72 h treatment with concentrations of DDT (). (B) Gene expression differences following 24 h treatment with DDT concentration (). A statistical difference between a treatment group and control is denoted as * as measured using a one-way ANOVA. (C) Representative staining of APP protein levels (scale bar: ). (D) Quantification of APP immunocytochemistry from 50–60 cells/well, calculated as the ratio of control (). (E) Quantification of released into the media from SH-SY5Y cells (). A significant difference between DDT alone compared with control (indicated by *), and (indicated by #) was measured using a two-way ANOVA. indicates the presence or absence of either DDT or TTX in the treatment. All data are presented as means, and error bars as ; * and # indicate . Summary data are reported in Tables S3–S6 for panels A, B, D, and E, respectively. Note: AD, Alzheimer’s disease; APOE, apolipoprotein E; APP, amyloid precursor protein; ANOVA, analysis of variance; BACE1, beta-secretase 1; DAPI, 4′,6-diamidino-2-phenylindole; DDT, dichlorodiphenyltrichloroethane; MAP2, microtubule-associated protein 2; NEP, neprilysin; PSEN1, presenilin-1; SD, standard deviation; TTX, tetrodotoxin.

We next measured the ability of DDT to modulate protein levels of APP in the presence and absence of the sodium channel antagonist TTX, given that sodium channels are the target of DDT-mediated neurotoxicity[46-48] (Figure 1C). Two-way ANOVA revealed a significant interaction between DDT treatment and TTX administration (, ), as well as independent main effects of DDT (, ), and TTX (, ) (Figure 1D; Table S5). Bonferroni post hoc tests revealed a significantly higher APP intensity in staining between the control group and DDT-treated cells (, , ) (Figure 1D). This effect was abolished when cells were pretreated with TTX (, , ) (Figure 1D). We also measured significantly higher secreted levels of (Figure 1E; Table S6) following DDT treatment after 24 h. Two-way ANOVA revealed a significant interaction effect (, ), and significant main effects for TTX (, ) and DDT treatment (, ). Inclusion of TTX showed significantly lower levels compared with SH-SY5Y cells exposed to DDT (, , ) (Figure 1E).

APP Expression and Protein Levels Post DDT Exposure in Primary Mouse Hippocampal and Cortical Neurons

To confirm the effects of DDT on AD markers in primary neurons, we cultured hippocampal and cortical neurons isolated from PND 0–1 pups of mixed-sex C57BL/6J mice. Cytotoxicity of DDT and cell viability at lower () DDT concentrations were confirmed by SYTOX assay in the primary hippocampal neurons after 72 h of exposure (Figure 2A; Table S7). Because DDT was the highest concentration that did not induce significant cell death, experiments in primary neuronal cells were conducted with this concentration. Following 24 h treatment with DDT, we found significantly higher App (, , ) mRNA levels in primary hippocampal neurons (Figure 2B; Table S8). In addition, immunofluorescent staining demonstrated greater intensity for APP in these cells following DDT treatment, which was abolished by the addition of TTX in both hippocampal and cortical primary neurons (Figure 2C). Two-way ANOVA revealed a significant interaction between DDT and TTX administration in the hippocampal neurons (, ) (Figure 2D; Table S9). APP levels were significantly higher in DDT-treated cells (, ) compared with control. Furthermore, this effect was suppressed in the presence of TTX (, ) (Figure 2D).

Figure 2.

APP mRNA and protein levels in primary mouse hippocampal and cortical neurons following DDT treatments. (A) Sytox assay (percentage cell death) data after 72 h treatment with concentrations of DDT in primary hippocampal neurons () as determined by one-way ANOVA. (B) App gene expression in primary hippocampal neurons following DDT treatment for 24 h () as determined by Student’s unpaired -test. (C) Representative staining of APP in primary hippocampal neurons isolated from C57BL/6J pups. (D) Quantification of APP immunocytochemistry from 60–90 cells/well in primary hippocampal neurons, calculated as the ratio of control () as determined by a two-way ANOVA. (E) Sytox assay (percentage cell death) data after 72 h treatment with concentrations of DDT in primary cortical neurons () as determined by one-way ANOVA. (F) App gene expression in primary cortical neurons following DDT treatment for 24 h () as determined by Student’s unpaired -test. (G) Representative staining of APP in primary cortical neurons isolated from C57BL/6J pups. (H) Quantification of APP immunocytochemistry from 60–90 cells/well in primary cortical neurons, calculated as the ratio of control () as determined by a two-way ANOVA. All data are presented as means, and error bars as . A significant difference between DDT alone compared with control is indicated by *, and is indicated by #. * and # indicate . indicates the presence or absence of either DDT or TTX in the treatment. Summary data are reported in Tables S7–S12 for panels A, B, D, E, F, and H, respectively. Note: APP, amyloid precursor protein; ANOVA, analysis of variance; DDT, dichlorodiphenyltrichloroethane; MAP2, microtubule-associated protein 2; SD, standard deviation; TTX, tetrodotoxin.

Next, we assessed the cytotoxicity of DDT ( for 72 h) in primary cortical neurons and identified DDT as the highest concentration that did not induce significant cell death (Figure 2E; Table S10). Similar to hippocampal neurons, DDT treatment resulted in higher App mRNA levels in primary cortical neurons (, , ) (Figure 2F; Table S11). Immunofluorescent staining showed greater intensity for APP in DDT-treated cortical neurons (Figure 2G). Two-way ANOVA indicated an interaction between DDT and TTX treatments (, ) (Figure 2H; Table S12). Significantly higher APP levels were observed in cortical neurons exposed to DDT compared with control (, ). Pretreatment of neurons with TTX suppressed this effect significantly (, ) (Figure 2H).

APP Expression and Protein Levels in Wild-Type Mice following Exposure to DDT

To confirm that systemic DDT exposure similarly produced these effects in APP-related genes, we exposed C57BL/6J mice to DDT every 3 d for 30 d. This exposure paradigm produced similar changes in gene expression as observed in vitro, although regional variations were observed. In DDT-treated males, hippocampal mRNA expressions were significantly higher for App (, , ), Psen1 (, , ), Nep (, , ) and Apoe (, , ) (Figure 3A, left; Table S13). Similarly, in DDT-treated females, hippocampal mRNA levels were significantly higher for App (, , ) and Nep (, , ); however, Apoe expression was significantly lower compared with control (, , ) (Figure 3A, right; Table S14). Furthermore, we assessed protein levels of APP in the hippocampus of male and female DDT-treated C57BL/6J mice. Using western immunoblotting, 70% (, ) higher APP levels were observed in the hippocampus of males when normalized to (, , ) (Figure 3B, left; Table S15). Similarly, in the hippocampus of female mice, (, ) higher APP levels were observed when normalized to (, , ) (Figure 3B, right; Table S16). Full blots are provided in Figure S1. APP was also visualized by IHC staining in the CA1 and CA3 hippocampal subregions of the male (Figure 3C, left) and female (Figure 3C, right) C57BL/6J mice.

Figure 3.

Effect of chronic DDT exposure ( every 3 d, 14 wk of age for 30 d) on (A,B) APP and AD-related genes and (C–F) APP protein in wild-type C57BL/6J mice. (A) Gene expression differences for App and metabolizing genes in the hippocampus of C57BL/6J male and female mice (). (B) Representative western blot and quantification for APP in the hippocampus of male and female mice (). (C) Representative staining of APP in the hippocampal CA1 and CA3 subregion of male and female mice. (D) Gene expression for App and metabolizing genes in the frontal cortex of C57BL/6J male and female mice (). (E) Representative western blot and quantification for APP in the frontal cortex of male and female mice (). (F) Representative staining of APP in the piriform cortex and layer V of the frontal cortex of male and female mice. Scale bar: . All data are presented as means, and error bars as ; * indicates (compared with control as assessed by Student’s unpaired -test). Summary data are reported in Tables S13–S20 for panels A, B, D, and E, respectively. Note: APP, amyloid precursor protein; DDT, dichlorodiphenyltrichloroethane.

In the frontal cortex of C57BL/6J male mice, no differences in gene expression were observed between DDT and control groups (Figure 3D, left; Table S17). In contrast, in the frontal cortex of DDT-treated females, we observed significantly lower expression of App (, , ), and Bace1 (, , ) (Figure 3D, right; Table S18). Western immunoblotting indicated 30% higher APP levels in the frontal cortex of males when normalized to (, , ) (Figure 3E, left; Table S19). In the frontal cortex of female mice, 35% (, ) higher APP levels were observed when normalized to (, , ) (Figure 3E, right; Table S20). Full blots are provided in Figure S2. In addition, APP was also visualized by IHC staining in the frontal cortex of the male (Figure 3F, left) and female (Figure 3F, right) C57BL/6J mice.

Levels in Transgenic Flies and Mice following Exposure to DDT

To extend these mechanisms to an orthologous model of regulation, we used two transgenic fly strains, , which overexpresses in the eye and exhibits a rough eye phenotype,[28] and , which overexpresses peptides in the CNS and exhibits a short life span phenotype.[32] Kaplan–Meier survival plots were generated for both fly strains (Figure 4A,B; Tables S21 and S22). Exposure to DDT significantly affected survival in (, ) by day 21 (Figure 4A). flies exposed to DDT had a significantly shorter life span compared with the vehicle control group (, ) (Figure 4B). Furthermore, we examined the effect of DDT on accumulation in both strains. In the flies, we classified the rough eye phenotype based on the level of severity, as exemplified by the arrows (Figure 4C). The number of flies exhibiting the severe phenotype from wk 0 to wk 2 was approximately doubled in flies exposed to DDT compared with control (, ) (Figure 4D; Table S23). To directly visualize the effect of DDT exposure on accumulation in the CNS, and the control HR strain (which does not express ) were treated with either 0.01% ethanol or DDT. The HR control strain exhibited minimal immunoreactivity, as expected, whereas the strain displayed greater fluorescence in both the control and DDT-exposed group (Figure 4E). This fluorescence appeared to be primarily localized to the mushroom bodies, the area of the CNS where is expressed in the transgenic flies. DDT-exposed flies had (, ) higher staining compared with controls (, , ) (Figure 4F; Table S24).

Figure 4.

Effect of DDT exposure on (A,B) survival, (C,D) rough eye phenotype, and (E,F) deposition in transgenic fly models of AD. (A) Kaplan–Meier survival plots for fly strain (expressing eye ) after exposure to ethanol (control) or DDT. (B) Kaplan–Meier survival plots for fly strain (expressing brain ) after exposure to ethanol (control) or DDT. experiments, each experiment consisting of flies. (C) Representative images of the rough eye phenotype exhibited by strain. Arrows indicate accumulation in the eye. (D) Quantification of severe rough eye phenotype of flies following exposure to DDT. For (A–D), experiments, and each experiment consisted of 100 individuals. (E) Representative staining of in the CNS of HR and flies (). Arrows indicate accumulation. (F) Quantification of staining of in the CNS of flies (). All data are presented as means, and error bars as ; * indicates compared with control as determined by Student’s unpaired -test. Summary data are reported in Tables S21–S24 for panels A, B, D, and F, respectively. Note: AD, Alzheimer’s disease; CNS, central nervous system; DDT, dichlorodiphenyltrichloroethane; HR, Hikone-R strain; SD, standard deviation.

The 3xTG-AD mouse model of AD expresses three human genetic mutations—PSEN1 (M146B), APP (APPswe KM670/671NL), and MAPT (P301L)—which lead to the development of and tau pathologies,[49,50] specifically plaque-like pathology and phosphorylated tau, particularly in female mice.[51,52] Mice were dosed with DDT ( every 3 d) starting at 9 months of age, when pathology begins to develop, until 12 months of age, when plaque pathology is readily observed in female transgenic mice.[52] Two-way ANOVA revealed significant main effects in cortical levels of for sex (, ) and treatment (, ) (Figure 5A, left; Table S25). Cortical levels were significantly higher by 20% (, ) in the female treatment group compared with the female control group (, , ) (Figure 5A, left). Two-way ANOVA also revealed significant main effects in cortical levels of for sex (, ) and treatment (, ), as well as an interaction effect (, ) (Figure 5A, middle; Table S26). A Bonferroni post hoc test revealed higher levels by 19% (, ) in females exposed to DDT compared with controls (, , ) (Figure 5A, middle). Measurement of oligomeric demonstrated similar findings, with significantly higher levels in the cortex (, , ) (Figure 5A, right; Table S27) of 3xTG-AD mice compared with vehicle-treated 3xTG-AD mice.

Figure 5.

(A,B) levels and (C) plaque staining in the cortex and hippocampus of 3xTG-AD mice following DDT exposure ( every 3 d, 9 months of age until 12 months of age). Quantification of (left), (middle), and oligomeric (right) in 3xTG-AD mice in the (A) cortex and (B) hippocampus. Only female samples were assayed for oligomeric . All data are presented as means, and error bars as ; * denotes a significant difference () between control and treatment groups, and the letter A denotes a significant difference between sex for the same treatment (). and levels were analyzed by two-way ANOVA, whereas levels were analyzed by Student’s unpaired -test. (C) Representative staining of in the cortex, CA1 of the hippocampus, and subiculum of male and female 3xTG-AD animals (Scale bar: ). Summary data are reported in Tables S25–S30 for (A) and (B), respectively. Note: ANOVA, analysis of variance; DDT, dichlorodiphenyltrichloroethane; SD, standard deviation.

In the hippocampus of 3xTG-AD mice, two-way ANOVA revealed significant main effects for sex () in levels (, ) (Figure 5B, left; Table S28). A significant main effect for sex (, ) and treatment (, ), as well as an interaction effect (, ), were observed for hippocampal levels (Figure 5B, middle; Table S29). levels were significantly higher in the female DDT-exposed group by 39% (, ) (, , ) compared with control. Measurement of oligomeric demonstrated significantly higher levels in the hippocampus (, , ) (Figure 5B, right; Table S30) of 3xTG-AD mice compared with vehicle-treated 3xTG-AD mice.

To confirm the quantitative measurements IHC staining for was performed in the CA1 subfield of the hippocampus, subiculum, and cortex (Figure 5C). Female 3xTG-AD mice demonstrated more prominent staining than males, which was exacerbated in the DDT-treated females, particularly in the CA1 and subiculum. Analysis of the plaque area in a limited number of samples supported the quantitative findings, with higher plaque area in all three regions of the DDT-treated female 3xTG-AD mice compared with vehicle-treated 3xTG-AD female mice (Figure S3; Table S31). Higher levels of and were detected in multiple regions of DDT-treated females. Furthermore, consistent with our in vitro and in vivo data from wild-type mice, we found that APP levels were significantly higher in the cortex and hippocampus of DDT-treated 3xTG-AD mice (Figure S4; Table S32). In addition, in line with our hippocampal gene expression data from wild-type mice, we found that APOE levels were significantly higher in the hippocampus of DDT-treated 3xTG-AD females by immunofluorescent staining (Figure S5; Table S33).

Synaptic Markers and Hippocampal Volume of Control and DDT-Exposed 3xTG-AD Female Mice

Because significant differences of were observed in 3xTG-AD female mice, we further quantified levels of synaptophysin and PSD95 by western immunoblot and normalized to in the cortex and hippocampus. These synaptic markers were investigated to determine whether exposure to DDT affected other end points that are characteristic of AD and the 3xTG-AD model.[51,52] In the cortical region of DDT-treated 3xTG-AD females, synaptophysin was 15% lower (, ) (, , ) (Figure 6A, top; Table S34) and PSD95 was (, ) (, , ) lower compared with control females (Figure 6A, bottom; Table S35). However, neither marker was affected in the hippocampus of female 3xTG-AD mice exposed for 3 months to DDT (Figure 6B; Tables S36 and S37). Full blots for these markers are provided in Figures S6 and S7.

Figure 6.

(A,B) Synaptic protein levels and (C) hippocampal volume in female 3xTG-AD mice following DDT exposure ( every 3 d, 9 months of age until 12 months of age). (A) Cortical and (B) hippocampal levels of synaptophysin (top) and PSD95 (bottom) protein in female 3xTG-AD mice. Data are normalized to and presented as percentage control (). All data are presented as and * indicates as determined by Student’s unpaired -test (C) Hippocampal volume () by subregion in female 3xTG-AD mice with and without exposure to DDT (). For hippocampal volume, * and # indicate statistical significance compared with control and DDT-treated groups, respectively. Data are presented as ; * and # indicate as determined by two-way ANOVA. Summary data are reported in Tables S34–S37 for (A) and (B), respectively. Note: ANOVA, analysis of variance; DDT, dichlorodiphenyltrichloroethane; DG, dentate gyrus; HC, hippocampal; SD, standard deviation.

Raw hippocampal volumes (in micrometers cubed) and percentage of hippocampal volumes for the CA1, CA2, CA3, and DG subfields in control and DDT-treated 3xTG-AD female mice are reported in Table S38. A mixed-effects ANOVA of hippocampal volume in control vs. DDT-treated 3xTG-AD female mice revealed a significant main effect for region (, ) but no effect for treatment (, ) and interaction between treatment and region (, ) (Figure 6C). A Bonferroni post hoc test found that CA1, CA3, and DG hippocampal volumes were significantly greater than CA2 in respective treatment groups ().

Discussion

Epidemiological studies from our laboratory,[12] as well as from other groups, support the potential involvement of DDT exposure as a risk factor for cognitive dysfunction and AD.[13-15] Although these studies provide evidence for an association between DDT and AD risk, they do not offer a potential mechanism by which this may occur. The only studies that examined the role of DDT in AD in vitro reported a significant increase in APP and BACE1 in H4 neuroglioma cells overexpressing APP[55] and our previous report demonstrated higher levels of APP in SH-SY5Y cells treated with DDT or DDE.[12] We extended our previous finding to demonstrate that DDT-exposed models had higher levels of the mRNA encoding APP, its cleavage enzymes, and its pathological products. Importantly, this activity appears to be mediated by the interaction of DDT with voltage-gated sodium channels (VGSCs), given that inclusion of the VGSC-blocker TTX blocked the effects of DDT on APP levels and secretion. Previously, other groups have established a role for electrical depolarization and increased neuronal activity in inducing APP and release, both of which were blocked with the inclusion of TTX.[56,57] Similarly, optogenetic stimulation in transgenic mice has been reported to increase levels of in the interstitial fluid (ISF) as measured by microdialysis.[58] Further, network hyperexcitability, possibly resulting from sodium channel disruption of inhibitory interneurons, is emerging as a potential contributor to cognitive deficits in AD.[59-61] The role of disrupted excitability has also been demonstrated in animal models of AD, where spontaneous seizures are commonly observed and reported.[62] These results provide, to our knowledge, the first mechanistic linkage between DDT exposure and , by implicating DDT interaction with the sodium channel.

In addition to the effects of DDT on APP levels and release, we also observed differences in expression of genes regulating amyloid metabolism, including higher levels of NEP mRNA, both in vitro and in vivo. Increased NEP levels have been found in the AD brain, with levels relative to disease severity.[63] Although the mechanism for these observations has not been determined, it may be due to a compensatory action after prolonged increases in both APP and . However, we observed contrasting effects of DDT on App gene expression and protein levels in the C57BL/6J female frontal cortex (Figure 3D,E), in that App gene expression was significantly lower, whereas immunoblotting revealed higher APP protein levels. It is important to note that these data were collected from a single time point after DDT exposure. Gene and protein levels often do not concur as a result of the dynamic balance of transcription, processing, and decay of mRNAs to the translation, localization, modification, turnover, and degradation of the proteins.[64] Further acute, subchronic, and chronic studies outlining the gene expression and protein levels of APP holoprotein and levels post DDT exposure are required to highlight brain region–specific trajectory of APP.

Of the target genes in amyloid processing and transport, Apoe had the greatest difference in expression (-fold) in the hippocampus of male C57BL/6J mice. Contrastingly, in the hippocampus of female C57BL/6J mice, a small, yet significant, down-regulation of Apoe gene expression was observed, which may be attributed to different APOE metabolism rates in males and females.[65] Furthermore, in the 3xTG-AD transgenic mouse model, we found that APOE protein levels were significantly higher as a consequence of DDT exposure in the hippocampus of female mice. In addition, immunofluorescent staining in the frontal cortex showed greater levels of APOE, including in many cells that were not GFAP positive, although this requires replication with additional animals (Figure S5). Evidence from single-cell RNA sequencing studies indicate brain region–specific differential expression of Apoe in microglia, astrocytes, and neurons.[66,67] In the brain, although APOE is predominantly produced by astrocytes and microglia,[68] previous studies demonstrated that APOE was up-regulated in neurons following neuronal injury or stress and that higher APOE levels were associated with production and early seeding of [69,70] Further, higher levels of APOE expression are widely recognized as a marker of the neurodegenerative phenotype of microglia,[71] including disease associated microglia (DAM)[72] and microglial neurodegenerative phenotype (MGnD).[73] Furthermore, microglial APOE has been associated with plaques providing evidence for the potential role of microglia in amyloid pathology in AD.[74] Thus additional studies will be required to determine the cell type and role of APOE in the effects of DDT.

A caveat of these experiments is that both C57BL/6J and 3xTG-AD mice have the mouse Apoe gene, which is known to have differential effects on AD-relevant pathways, including pathology, compared with human APOE gene.[75] Future studies using male and female humanized and mice will be crucial for understanding the association between DDT and APOE. This is particularly important because, in our previous epidemiological study, we observed that individuals with an allele and who were in the top tertile of serum DDE levels displayed significantly lower mini-mental state exam scores than either or top tertile DDE alone, demonstrating a significant gene–environment interaction.[12] Because carriers have been shown to have significantly higher amyloid load than carriers,[70] and has been reported to stimulate APP expression and seeding more than ,[76-78] increased levels of APOE by DDT could lead to enhanced plaque formation.

To determine the effects of systemic DDT on accumulation, we used transgenic flies and mice that exhibit amyloid pathologies. In these experiments, we observed conservation of the effects of DDT on APP in models where pathogenic could be measured along with plaques. Using transgenic fly strains, we found that accumulation in the eye and CNS was higher in DDT-treated groups (Figure 4C–F). In 3xTG-AD mice, a well-characterized genetic model of AD, we were able to assess a number of key questions with regard to DDT exposure. First, elevated levels and plaque-like formation were observed following DDT exposure, particularly in female mice. This model is unique in that the female mice exhibit clear differences in pathology as compared with the males, with the females exhibiting greater levels of in the hippocampus and cortex (Figure 5A,B). These data are consistent with those previously reported, indicating higher peptide levels in the brains of female 3xTG-AD mice compared with male mice at the time point we measured.[52] It remains to be determined whether a longer exposure period and aging of the 3xTG-AD male mice would reveal effects on . Another unique opportunity provided by the 3xTG-AD model is that we were able to measure human levels, as detected by a species-specific antibody (6E10) used by the MSD system, the same platform as some human biomarker studies,[79-81] which should aid in translational research. However, future work is required to determine whether female susceptibility combined with DDT exposure is unique to the 3xTG-AD animal model, or if it occurs across different models of AD, including those with prominent tau pathology.

Synaptic deficits, cortical thinning and reduced hippocampal volumes are prominent features of AD.[82] In the 3xTG-AD model, synaptic deficits, as measured by decreases in PSD95 and synaptophysin are observed without a corresponding loss in hippocampal neurons or volume.[83] Here, we measured protein levels of synaptophysin and PSD95 in the hippocampus and cortex along with hippocampal volume to determine whether exposure to DDT was associated with additional changes consistent with AD pathogenesis. Our data provide evidence that exposure to DDT is associated with lower levels of PSD95 and synaptophysin in the cortex, but not in the hippocampus. Furthermore, we also report no association with exposure to DDT and any changes with hippocampal volume. Although we observed greater plaque formation in the hippocampus of DDT-treated 3xTG-AD mice, soluble species are thought to be the primary contributors to synaptic dysfunction in AD and the 3xTG-AD mice.[54] We observed higher levels of total soluble and oligomeric in the hippocampus and cortex of DDT-treated 3xTG-AD mice. The largest difference in levels was significantly higher in the cortex. This is potentially important, given that soluble has been shown to correlate highly with synaptic loss in postmortem brain samples of patients with AD[84] and N-methyl-d-aspartate–dependent decreases in PSD95.[85] However, the relationship between soluble and deficiency in synaptic markers following DDT exposure remains to be firmly established and further studies are needed to address the loss of synaptic markers in the cortex.

Our results are novel and comprehensive in that they have been replicated in a number of different model systems, from a widely used human neuroblastoma cell line, to primary neuronal cultures, and two in vivo transgenic systems. In all models, we demonstrated that DDT had significant impacts on the amyloid pathway by elevating App mRNA and, subsequently, protein levels. Moreover, using two transgenic systems, we have also shown the potential of DDT to exacerbate amyloid-like pathology, specifically levels of . Taken together, our data provide compelling evidence that DDT exposure affects the amyloid pathway. The level of DDT used in our in vivo and in vitro studies is biologically relevant and is in the range of what people were exposed to in the 1960s and 1970s. Specifically, average nonoccupational daily human consumption of DDT during this time was estimated to be between , which would be in an average human.[20,86] Using allometric scaling, which normalizes dose to body surface area,[87] the human equivalent dose administered to mice in this study, calculated as the mouse dose divided by 12.3, is estimated to be . This dose is particularly relevant considering the metabolism differences among the species, which likely influences the substantial difference in the lowest observed adverse effect level of mice ()[88] compared with humans ()[89] based on neurological end points.[20] This may potentially be attributed to rapid metabolism of DDT by rodents.[90] Furthermore, humans who were exposed occupationally in the United States in the 1940s–1970s and in areas where DDT has been deployed to combat malaria, were potentially exposed to much higher levels of DDT[91,92] and are now beginning to develop, or are well into the range of ages with higher risk for developing, AD.

It is becoming widely accepted that AD presents as a heterogeneous disease. This includes many different pathologies, such as those that may be or tau driven, those involving chronic neuroinflammation, and those that involve a complex combination of these factors, as well as many other genetic and environmental factors.[4,93] Thus, the initiating events for different individuals likely vary based on genetic and environmental risk factors. Indeed, based on our previous human data,[12] the impact of APOE genotype and DDT exposure on AD pathology potentially represents a crucial gene–environment interaction, and it will be important to determine the mechanism(s) responsible for this interaction. Taken together, our findings may have significant relevance for personalized medicine approaches and early identification of those that may be at particular risk for AD. Identification of specific environmental risk factors, such as pesticide exposure, and their subsequent mechanisms of action and interaction with genetic risk factors may lead to additional diagnostic tests and optimized individual treatment approaches.

Click here for additional data file.

Click here for additional data file.