Sex and Gender Differences in Alzheimer's Disease

Sex and Gender Differences in Alzheimer’s Disease: The Women’s Brain Project offers for the first time a critical overview of the evidence documenting sex and gender differences in Alzheimer’s disease neurobiology, biomarkers, clinical presentation, treatment, clinical trials and their outcomes, and socioeconomic impact on both patients and caregivers. This knowledge is crucial for clinical development, digital health solutions, as well as social and psychological support to Alzheimer’s disease families, in the frame of a precision medicine approach to Alzheimer’s disease.This book brings together up-to-date findings from a variety of experts, covering basic neuroscience, epidemiology, diagnosis, treatment, clinical trials development, socioeconomic factors, and psychosocial support. Alzheimer’s disease, the most common form of dementia, remains an unmet medical need for the planet. Wide interpersonal variability in disease onset, presentation, and biomarker profile make Alzheimer’s a clinical challenge to neuroscientists, clinicians, and drug developers alike, resulting in huge management costs for health systems and society. Not only do women represent the majority of Alzheimer’s disease patients, but they also represent two-thirds of caregivers. Understanding sex and gender differences in Alzheimer’s disease will lead to novel insights into disease mechanisms, and will be crucial for personalized disease management strategies and solutions, involving both the patient and their family.

Endorsements/Reviews:

"There is a clear sex and gender gap in outcomes for brain health disorders like Alzheimer’s disease, with strikingly negative outcomes for women. This understanding calls for a more systematic way of approaching this issue of inequality. This book effectively highlights and frames inequalities in all areas across the translational spectrum from bench-to-bedside and from boardroom-to-policy and economics. Closing the Brain Health Gap will help economies create recovery and prepare our systems for future global shocks." Harris A. Eyre MBBS, PhD, co-lead, Neuroscience-inspired Policy Initiative, OECD and PRODEO Institute. Instructor in Brain Health Diplomacy, Global Brain Health Institute, UCSF and TCD.

"Sex and Gender Differences in Alzheimer's disease is the most important title to emerge on Alzheimer's disease in recent years.This comprehensive, multidisciplinary book is a must read for anyone with a serious interest in dementia prevention, diagnosis, treatment, care, cure and research. Precision medicine is the future of healthcare and this book represents an incredible and necessary resource to guide practice, policy and research in light of the fact that Alzheimer's disease disproportionately affects women. The combination of contributions from the most eminent experts and the most up-to-date research makes this an invaluable resource for clinicians, care providers, academics, researchers and policy makers. Given the complex nature of dementia and the multiple factors that influence risk and disease trajectory the scope of the book is both impressive and important covering sex differences in neurobiological processes, sex and gender differences in clinical aspects and gender differences linked to socioeconomic factors relevant to Alzheimer's disease. If you work in Alzheimer's disease, or indeed other dementias, then Sex and Gender Differences in Alzheimer's disease is a must have for your bookshelf." -- Sabina Brennan, PhD., C.Psychol.,PsSI., National representative for Ireland on Alzheimer Disease International's Medical and Scientific Advisory Panel

• Language: English
• Publisher: Academic Press
• Release date: Jul 23, 2021
• ISBN: 9780128193457

Book preview

Section 1

Sex differences in fundamental neurobiological processes that are relevant to Alzheimer’s disease

Introduction

Maria Teresa Ferrettia; Annemarie Schumacher Dimecha,b; Antonella Santuccione Chadhaa, a Women’s Brain Project, Guntershausen, Switzerland, b Program Manager, Palliative Care, University of Lucerne, Lucerne, Switzerland

As every cell has a sex, differences can be found in many fundamental cellular mechanisms throughout the body, as well as the brain. Such differences, which we are just starting to uncover, underlie the female- and male-specific presentation of diseases and deserve attention. For this reason, the book will start with three chapters that describe basic sex differences observed in the field of neurobiology and that could have an impact on the disease of Alzheimer’s and dementia at large.

In Chapter 1, Ippati, Ittner and Ke provide an overview of sex differences in animal models widely used to study Alzheimer’s disease; the knowledge of such differences is crucial, especially when using such models for drug development.

In Chapter 2, Lee, Puri and Galea describe the effects of sexual hormones on neurogenesis and synaptic plasticity. Sex differences in these fundamental processes are likely to impact the development of Alzheimer’s pathology, especially during hormonal changes that characterize the menopausal/andropausal state.

Finally, in Chapter 3, Delage, Rendina, Malacon, Tremblay and Bilbo summarize the emerging field of sex differences in neuroimmunology, with a particular emphasis on microglial cells—the brain’s resident immune cells.

These three chapters focus mostly on biological sex and only refer to gender-related aspects in specific cases.

Chapter 1: Sex differences in Alzheimer’s disease animal models

Stefania Ippatia; Lars Matthias Ittnerb; Yazi Diana Keb    a San Raffaele Scientific Institute, San Raffaele Hospital, Milan, Italy

b Dementia Research Centre, Department of Biomedical Sciences, Faculty of Medicine Health and Human Sciences, Macquarie University, Sydney, NSW, Australia

Abstract

Alzheimer’s disease (AD) is the most prevalent of all neurodegenerative disorders. Distinct protein inclusions, amyloid-β (Aβ) plaques, and neurofibrillary tangles (NFTs) characterize the AD brain. Several epidemiological studies have shown dissimilarities in AD when sex is considered, which has driven the scientific community, over the years, to consider sex as a determinant factor in AD pathology. Similarly, studies in animal models have been moving forward to investigate the genetic and metabolic basis of sex differences in AD for the development of possible new sex-specific therapeutic strategies. However, sex differences are still poorly understood, and preclinical studies should increasingly consider the relevance of such disparities. So far, most findings have pointed to sex differences in murine transgenic genetic backgrounds and dependently on age. Other studies have found sex-dependent AD pathological hallmarks within different rodent models. Hence, studies aiming to understand AD pathogenesis, diagnostic and prognostic criteria, and response to therapy must include the study of sex. In this chapter, we describe the state of the art behind the sex differences in AD so far, and discuss preclinical aspects considered in these studies.

Keywords

Alzheimer’s animal models; Transgenic animal models; Experimental medicine; Gender medicine; Sex

Overview of Alzheimer’s disease

Alzheimer’s disease (AD) is the most prevalent of all the neurodegenerative disorders characterized by a progressive loss of cognition. Distinct protein inclusions define most neurodegenerative diseases. In particular, the AD brain is characterized by two types of protein deposits, amyloid-β (Aβ) plaques and neurofibrillary tangles (NFTs), the latter formed by hyperphosphorylated forms of the microtubule-associated protein tau (Querfurth & LaFerla, 2010). Over recent decades, experiments on cellular and animal models have suggested that AD pathogenesis involves assemblies of Aβ peptides as a pathological trigger for what is commonly known as the amyloid cascade hypothesis, which includes numerous pathological cellular events such as neuritic injury, inflammatory responses, oxidative stress, formation of neurofibrillary tangles via tau, neuronal dysfunction, and finally cell death (Hardy & Selkoe, 2002). Hence, clinical trials have focused on drug and vaccine development against Aβ or tau pathology as therapeutic approaches to AD (Panza, Lozupone, Logroscino, & Imbimbo, 2019). However, no clinical benefits by targeting Aβ or tau in humans have been obtained as yet (Imbimbo, Ippati, & Watling, 2020).

AD animal models

Animals are important models for Alzheimer’s disease (AD) research. Over the past two decades, AD animal models have helped clarify the etiology of AD, testing therapeutic interventions, and validating hypotheses of mechanisms in the onset and progression of the disease. Mouse models are the most frequently used animal models of AD (Götz & Ittner, 2008). However, the pathophysiology observed in animal models may not necessarily reproduce all clinical findings observed in patients, which may explain why it is challenging to translate successful preclinical research in AD mouse models into clinical practice. The first transgenic mouse models for AD, established more than two decades ago, expressed human amyloid-β precursor protein (APP) carrying a single gene mutation associated with the inherited familial Alzheimer’s disease (FAD) and recapitulated the histopathological lesions that are found in diseased human brains (Götz, Bodea, & Goedert, 2018). Since then, researchers have applied more precise gene-editing technologies to introduce pathogenic mutations in AD animal models and to accomplish endogenous gene expression levels rather than unphysiological overexpression (Scearce-Levie, Sanchez, & Lewcock, 2020). However, while most AD mouse models display histopathological features of AD (plaques and NFTs), they do not present the extensive neurodegeneration observed in the brains of AD patients. Moreover, the reproducibility of animal model research findings across laboratories is critical for the successful translation of experimental therapies to humans. The main mouse models that we will consider in this chapter are those most commonly used in preclinical studies

Fig. 1

Fig. 1 The AD hallmarks displayed in (a) are worse in female compared to male mice in the listed mouse models; (b) worse in male mice compared to female mice in the listed mouse models; (c) worse in male rats compared to female rats in the STZ AD model.

Mutant APP mouse models

The majority of AD models consist of transgenic mice expressing mutated human genes that result in the formation of amyloid plaques (by the expression of human APP alone or in combination with human presenilin 1 (PSEN1)) and neurofibrillary tangles (by the expression of human microtubule-associated protein tau (MAPT)) (Onos, Sukoff Rizzo, Howell, & Sasner, 2016). One of the earliest mouse models of Alzheimer’s disease, the PDAPP line, expressing the human APP with the Indiana mutation V717F under control of the PDGF-β promoter, displayed human AD-like phenotype, including plaques in the hippocampus, gliosis, synaptic dysfunction, and cognitive impairment (Games et al., 1995). The Tg2576 line, one of the best-characterized and widely used mouse models of AD, expressing the human APP with the double Swedish mutation (APPK670N/M671L) driven by the hamster prion protein promoter (PrP), develops plaques in the frontal, temporal, and entorhinal cortices, hippocampus, and cerebellum. In addition, cerebral amyloid angiopathy (CAA), synaptic impairment, gliosis, and memory impairment are presented by this mouse model (Hsiao et al., 1996). APP23 mice also express K670N/M671L-mutant APP; however, they differ from Tg2576 mice by expressing the longer APP751 isoform driven by the neuronal Thy1 promoter. APP23 mice have more pronounced CAA, form compact plaques in comparison to the predominantly diffuse plaques found in Tg2576 mice, and have localized neurodegeneration that is not seen in the Tg2576 mice (Allué et al., 2016).

Mouse models with multiple FAD mutations

Transgenic mice with more severe pathology developed at a younger age have been developed by expressing multiple FAD-associated mutations together, such as the J20 line that expresses both the APP Swedish and Indiana mutations, and the APP/PS1 mouse, expressing both human APP (Swedish mutation) and PSEN1 (L166P mutation) transgenes under the control of the Thy1 promoter (Mucke et al., 2000). APP/PS1 transgenic mice are widely used in AD research and have been developed with various specific FAD mutations and promoters. The APP/PS1 5xFAD mouse model expresses the Swedish (K670N/M671L), London (V717I), and Florida (I716V) APP mutations, M146L/L286V-mutant PSEN1 (Drummond & Wisniewski, 2017). The expression of five FAD mutations results in early intraneuronal Aβ accumulation at 6 weeks, followed by plaque formation at 2 months.

Mouse models of tau pathology

Mutant tau mice that showed NFTs formations expressed human tau containing mutations associated with frontotemporal lobar degeneration (FTLD). The first NFT-developing mouse was achieved by expressing the familial FTDP-17 MAPT mutation P301L under the control of the mouse PrP (Lewis et al., 2000). Another line, TAU58/2, expressing the P301S tau mutation under the control of the Thy1 promoter, presented early-onset motor deficits, progressively hyperphosphorylated tau resulting in NFT formation throughout the brain, increasing with age, and axonal pathology similar to human FTLD-tau and AD (van Eersel et al., 2015). Furthermore, mouse models that display both plaques and tangles have been reported. These models are characterized by the expression of mutated forms of APP and MAPT with or without PSEN1 or PSEN2 to drive plaque and tangle formation in the same model. The most commonly used model is the 3xTg-AD line, which develops intraneuronal Aβ at 3–4 months, followed by plaque development at approximately 6 months in the cortex and hippocampus. NFTs form at approximately 12 months, initially in CA1 and then in the cortex. To avoid transgene overexpression, gene targeting by homology-directed recombination (HDR) has been pursued to establish models, including ones in which AD-related genes are deleted (knockout (KO) models) or inserted (knock-in (KI) models) in endogenous murine gene loci. For example, in the APPNL-G-F knock-in model, APP is not overexpressed, avoiding potential artifacts. In this model, levels of pathogenic Aβ are elevated due to the combined effects of three mutations associated with FAD (Saito et al., 2014). As these models are only emerging, potential sex differences have yet to be fully established.

Other AD mutant models

Over the past decade, neuroinflammation has been a focus of AD studies (see Chapter 3). Brain immune cells are associated with major AD risk factors, such as apolipoprotein E (APOE), a protein that is mainly produced by brain astrocytes, and TREM2, which is expressed by peripheral macrophages and brain microglial cells (Guerreiro et al., 2013; Liu, Kanekiyo, Xu, & Bu, 2013). In particular, in population-based studies, the epsilon-4 allele of APOE was reported to be a weaker predictor among African American (odds ratio (OR) 5.7) and Hispanic (OR 2.2) populations and instead a stronger predictor in Japanese people (OR 33.1) compared with white individuals (OR 12.5) (Farrer, 1997). The p.R47H variant of TREM2 was also found to confer a significant risk of AD in Iceland (OR 2.92) (Jonsson et al., 2013), and the same variant has been associated with both early and late-onset AD in the Spanish population (Benitez et al., 2013). Therefore, double/triple-mutant mouse lines have been generated by crossing FAD mice with APOE4 KI mice, which carry a humanized APOE4 gene, and with Trem2 KO or KI lines.

Transgenic rat models, overexpressing various FAD mutations, have been generated (Cohen et al., 2013; Leon et al., 2010). Basic neuropathological and behavioral characterization showed similar phenotypes in males and females McGill-R-Thy1-APP rats (Leon et al., 2010); minor differences in Morris Water Maze (Berkowitz, Harvey, Drake, Thompson, & Clark, 2018) and buried food task (Saré et al., 2020) performance have been described in the Tg-F344 model. Much less commonly used rat AD models include the use of vectors in which AD-related genes are selectively expressed in brain regions relevant to the disease and chemically induced models using intracranial streptozotocin administration. In the following chapter, we will discuss AD rodent models in which the phenotypic presentations have been extensively investigated in the light of sex differences. To our knowledge, mouse and rat models of Aβ infusions have not been characterized for sex differences.

Evidence for sex differences in AD animal models

The triple transgenic AD mouse model: 3xTg-AD

Age-dependent behavioral sex differences

Most animal studies that have addressed sex differences in AD have used transgenic mouse models. The widely used 3xTg-AD mouse, which expresses transgenes carrying three mutations associated with FAD (APP KM670/671NL, MAPT P301L, and PSEN1 M146V), has been extensively characterized for sex differences. One of the first studies conducted by LaFerla’s laboratory investigated how sex influenced cognitive and neuropathological phenotypes and found that male and female 3xTg-AD mice show comparable behavioral impairments in the Morris water maze (MWM) inhibitory avoidance (IA) tests at 4 months (Clinton et al., 2007). Similarly, 3xTg-AD animals tested at 6 months of age for spatial cognition and memory displayed no sex differences (Giménez-Llort et al., 2010). However, Clinton and colleagues also found that female 3xTg-AD mice showed more significant deficits from 4 to 12 months than male 3xTg-AD mice in stressful behavioral tasks such as MWM, despite histological quantification of Aβ and tau brain pathology being comparable between sexes. Female 6-month-old 3xTg-AD mice were found to be impaired at spatial reorientation and at use of distal cues (Stimmell et al., 2019), while 6- and 12-month-old male 3xTg-AD mice were not impaired and presented with less brain pathology than females. Similarly, Yang et al. found that female 12-month-old 3xTg-AD mice displayed greater spatial cognitive deficits than males (Yang et al., 2018).

In the study from Clinton et al., sex differences were attributed to greater stress response in young female mice, also displayed by higher plasma corticosterone. Interestingly, the differential stress response between sexes was no longer apparent after 12 months of age, along with the disappearance of cognitive differences. In the same study, nonstressful conditions, such as the object recognition test, determined no behavioral sex differences across 3xTg-AD mice’s lifespan. Old age has also been found to influence major sex-dependent changes in social behavior of 3xTg-AD mice (Bories et al., 2012) associated with a significant increase in prefrontal cortex synaptic activity (mIPSC and mEPSC frequencies). Moreover, 3xTg-AD females were more disinhibited than their male counterparts at 12 months of age, while the opposite was true at 18 months. Interestingly, increased Aβ42 concentrations in the brain were associated with disinhibition in male 3xTg-AD mice, while an inverse association was observed in females.

In summary, according to the studies mentioned earlier, it is evident that female 3xTg-AD mice from 6 to 12 months are more affected by AD pathology than are male 3xTg-ADs, and that these variabilities appeared to be no longer present in mice after 12 months.

Sex differences in glucose homeostasis and immune function

Concerning age, Giménez-Llort et al. have shown that at early stages of the disease, female 3xTg-AD mice showed a poorer ability to maintain weight and glucose homeostasis, while increased peak blood glucose was observed in both sexes. Total weights of the thymus were higher in female than in male 3xTg-AD mice, and this sex difference was more obvious when relative weights were considered, suggesting possible differences in immune function between the sexes. Amyloid deposits appeared slightly larger in females than in males, in agreement with other observations in this mouse line (Carroll et al., 2010; Yang et al., 2018), while male 3xTg-AD mice displayed increased lipid peroxidation and derangement of the glutathione system.

Olfaction and hormonal-related factors

It has been suggested that sex steroid hormones may influence sex differences in Aβ pathology in 3xTg-AD mice during development (Carroll et al., 2010), as male 3xTg-AD mice that were demasculinized during early development exhibited significantly increased Aβ accumulation in adulthood, while female mice defeminized during early development exhibited a less severe Aβ pathology in adulthood. Sex-dependent behavioral assessment of olfaction has been evaluated in 6-month-old 3xTg-AD mice, using an olfactory detection task with ethyl acetate as odorant (Roddick, Roberts, Schellinck, & Brown, 2016). At the highest odor concentration, male and female 3xTg-AD mice did not differ in accuracy from their wildtype controls, while only female 3xTg-AD animals showed impairment at the lowest concentration of the odorant. This study also analyzed 5XFAD mice, expressing the human APP and PSEN1 transgenes with a total of five AD-linked mutations, and found that no olfactory deficits were present in either sex or compared to their wildtype controls.

Increased brain Aβ and tau pathology and neuroinflammation in female 3xTg-AD mice were associated with an altered PKA-CREB-MAPK signaling pathway, which is closely associated with synaptic plasticity and memory (Yang et al., 2018). In particular, expression levels of p-PKA and p-CREB in the hippocampus were markedly lower, and p38-MAPK was higher in 3xTg-AD mice, and especially in females. The authors suggested that impaired PKA-CREB-MAPK signaling might be induced by estrogen deficiency in 12-month-old female mice, and estrogen supplementation or gene therapy targeting PKA-CREB-MAPK signals could be beneficial in AD older females (see Chapter 2).

Lifespan

Studies have focused on sex differences in 3xTg-AD mice of various ages, underlining how age appears to have an appreciable role in determining sex differences in this mouse strain. It appears that female 3xTg-AD mice are generally more vulnerable to AD pathology from 4 up to 12 months, but on the contrary, and in line with a previous study (Rae & Brown, 2015), Kane and colleagues reported that male 3xTg-AD mice had a shorter lifespan than female 3xTg-AD mice (Kane et al., 2018). Moreover, male 3xTg-AD mice lived for a shorter time than male WT mice, but this genotype difference in lifespan was not seen in female mice. Additionally, both male 3xTg-AD and WT mice had higher frailty scores than the corresponding female groups, suggesting that these sex differences may be related to differences in genetic factors, epigenetic factors or immune system variables.

APP and PSEN1 transgenic mouse models: TgCRND8, APPswe/PSEN1dE9, APPPS1, and PS2APP/Trem2KO

In the TgCRND8 mouse line, an APP transgenic mouse model carrying the APP Swedish and Indiana mutations, male and female animals have been reported to exhibit adrenocortical hyperactivity, an endocrine hallmark of AD (Green, Billings, Roozendaal, McGaugh, & LaFerla, 2006), in a sex-specific manner (Touma et al., 2004). In 4-month-old TgCRND8 mice, the plasma corticosterone concentrations of both sexes were elevated; but whereas fecal corticosterone metabolites (CM) were present at 45 days in male mice, in females, they only appeared at 90 days, suggesting an involvement of adrenocortical activity in an age- and sex-dependent manner. However, adrenal tyrosine hydroxylase activity measured at 4 months showed no significant differences between the sexes. This study also found no sex differences in the number of amyloid plaques observed or in exploratory and anxiety-related behavior. Li and colleagues investigated sex differences in an APPswe/PSEN1dE9 mouse model containing the APP Swedish mutation and a knocked-in variant of the human PSEN1 (deltaE9) gene. They found that 6-month-old male APP/PS1 mice showed abnormal glucose and insulin tolerance and higher total cholesterol and triglyceride levels compared to female APP/PS1 mice. However, the amounts of Aβ in female APP/PS1 mice were significantly higher than in the males, and corresponded to reduced memory test performance of females as assessed in the MWM (Li et al., 2016). The authors concluded that, in this mouse model, Aβ deposits might affect cognitive function more than impaired insulin signaling and elevated plasma lipid levels.

Age- and sex-specific alterations in brain electrical network alterations have been described in APPswePS1dE9 mice (Papazoglou et al., 2016). Whereas male APPswePS1dE9 mice exhibit a reduction in theta waves during electroencephalography (EEG) recordings at the age of 14 weeks, which disappear as they age, female mice exhibit theta power reduction at 18 and 19 weeks of age. These findings are in line with a report from 2015 in which APPswePS1dE9 male and female mice showed significantly different brain deposition kinetics for Aβ (Ordóñez-Gutiérrez, Antón, & Wandosell, 2015). In particular, senile plaques in the cortex and hippocampus could be detected in 3-month-old male APPswePS1dE9 mice, but barely in females of the same age; however, starting from 9 months of age, amyloid levels in plasma increased among females but decreased among males. Aβ plaque deposition in the retina has been reported to appear earlier in aged female than male APPswePS1dE9 mice (Perez, Lumayag, Kovacs, Mufson, & Shunbin, 2009).

Another widely used AD mouse model is the APPPS1 line, expressing transgenes for both human APP bearing the Swedish mutation and human PSEN1 with an L166P mutation, both under the control of the murine neuronal Thy1 promoter (Radde et al., 2006). Several studies have focused on sex-dependent AD pathology using the APPPS1 line. Dodiya and colleagues investigated how antibiotic treatments resulted in microbiome changes in male and female APPPS1 mice (Dodiya et al., 2019). While postnatal antibiotic treatment determined similar changes in gut bacterial composition at 22 days of age in both sexes, long-term antibiotic cocktail treatment (ABX) resulted in sex-specific microbiome changes. Some bacterial clusters associated with gut tissue degradation and inflammatory activation (e.g., Akkermansia muciniphila and an Allobaculum spp.) were significantly enriched in females compared to males, along with proinflammatory pathway activation, such as the bacterial secretion system, bacterial toxins, lipopolysaccharide biosynthesis, and lipopolysaccharide synthesis protein. Moreover, ABX-treated male APPPS1 mice exhibited elevated levels of antiinflammatory/neuroprotective factors and, in contrast, ABX-treated female mice showed upregulation of proinflammatory cytokines/chemokines. Furthermore, long-term ABX treatment resulted in reduced Aβ deposition, neuroinflammation, and antiinflammatory transcriptional signatures only in male APPPS1 mice.

Various studies have found that the mood stabilizer valproic acid (VPA), traditionally used to treat bipolar disorder, may have therapeutic potential in other central nervous system diseases (Williams, Cheng, Mudge, & Harwood, 2002). Evidence indicates that there are neurotrophic and neuroprotective sex-related effects of VPA induced on AD pathology in APPPS1 mice. VPA treatment was found to relieve anxiety-related behavior in male APPPS1 mice, whereas no significant beneficial effects were found in female mice between the VPA- and saline-treated groups. The same results were obtained during MWM testing, where VPA-treated APPPS1 male mice exhibited reduced escape latency compared to treated females. Moreover, VPA modified the synaptic structure of male mice, while there was significantly reduced Aβ brain burden and neuronal cell death in both sexes. Genetic variants of TREM2, a gene with expression in brain microglia, have been shown to increase the risk of developing late-onset AD (Ulland & Colonna, 2018). Since human TREM2 variants were linked to a reduction of TREM2 function (Cheng-Hathaway et al., 2018), Trem2KO mice have been crossed with APP transgenic mice to study the effects of loss of TREM2 function in the context of amyloidosis and tauopathy. In a recent study, PSEN2APP/Trem2KO female mice have been reported to accumulate Aβ pathology more rapidly than males, while later, as the mice aged, plaque accumulation was reduced in both female and male PSEN2APP/Trem2KO mice compared to age-matched TREM2WT, and neuroinflammation was similar in both sexes (Meilandt et al., 2020).

ApoE transgenic mouse models: ApoE4/3xTg, 5xFAD/ApoE3, and 5xFAD/ApoE4

The ApoE KI strain of mice have been established to investigate the role of Apolipoprotein E4 (ApoE4), the most prevalent genetic risk factor of AD (Liu et al., 2013). Being a potential therapeutic target for AD, mechanisms involved in the interaction between ApoE4 and sex have been studied repeatedly. Hou and colleagues have evaluated sex-dependent ApoE4 effects on learning, memory, and AD histopathology using ApoE4/3xTg mice. Their findings from behavioral and spatial exploration testing support the hypothesis that ApoE4 contributes to learning and memory deficits in 10-month-old ApoE4/3xTg female mice, which may be associated with hormonal fluctuation starting around this time (Hou et al., 2015). Using mouse vaginal mucosa smears, it was confirmed that ApoE4/3xTg transgenic animals’ cycle length became irregular and prolonged (>  5 days) compared to the normal cycle lasting 4–5 days. Moreover, female mice carrying ApoE4 displayed spatial and memory impairment earlier than their male counterparts and showed more prominent AD pathology in the hippocampus. Importantly, female ApoE4/3xTg mice showed increased BACE1 enzymatic activity and elevated expression of BACE1 and of its transcription factor SP1.

In a more recent report, microglial interactions with amyloid plaques have been investigated in 5xFAD/ApoE3  +/+ (E3FAD) and 5xFAD/ApoE4  +/+ (E4FAD) mice, and in particular, how ApoE genotypes and sex influence neuropathology in these mouse models (Stephen et al., 2019). The authors found that the ApoE genotype was a significant factor, with higher plaque coverage in E3FAD males than in E4FAD males. Further, there was a significant effect on microglial engulfment of plaques, which was twofold greater in male E3FAD than in female E3FAD mice. Moreover, in male E3FAD, microglial process interactions were increased and were associated with reduced plaque size. Using quantitative confocal analysis to check for plaque morphology, it has been found that TREM2 expression, microglial plaque coverage, and compaction, considered a beneficial consequence of microglial interactions with plaques, was diminished by ApoE4 and by female sex in E4FAD mice. Moreover, ApoE4 genotype and female sex were associated with the highest amyloid burden.

Tau mouse models: P301S, rTg4510, JNPL3

Tau P301S transgenic mice are one of the most widely used mouse models in AD research, and in recent years, have been considered for studies on sex- and age-related differences in neuropathology and behavior. Van Eersel and colleagues have investigated age and sex differences in the TAU58/2 line, a model of FTLD expressing the human 0N4R tau isoform with the P301S mutation, under the control of the murine Thy1.2 promoter. TAU58/2 mice presented early-onset motor deficits, axonal pathology, and NFT formation throughout the brain, with males displaying significantly more pronounced pathology than females (van Eersel et al., 2015). The authors found that the number of NFTs increased significantly in male TAU58/2 mice over time in the cortex, hippocampus, and brainstem. By 6 months of age, male TAU58/2 mice presented significant motor deficits, and at 10 months of age, both male and female TAU58/2 showed significant motor impairments. Furthermore, male TAU58/2 mice demonstrated greater amounts of phosphorylated tau-specific staining than female mice. Total soluble human tau levels in the cortex, hippocampus, and tau phosphorylated at Ser396/Ser404 (PHF-1) and Ser422 (pS422), and astrogliosis was significantly increased in male compared to female TAU58/2 mice. This corresponded to significantly higher amounts of insoluble tau in hippocampal extracts from male as compared to female TAU58/2 brains.

More recently, similar results have been obtained in P301S mice, although with a different background, and transgene driven by the murine prion protein instead (Sun et al., 2020). Male P301S mice undergo faster weight loss, more severe dyskinesia, and more severe memory dysfunction than female transgenic mice. Consistent with the sex differences in behavior and neuropathology, several plasma factors, including MIG, TNF-α, IL-10, and IL-13 exhibited specific changes in male P301S mice compared to females. In the same mouse line, microglial microRNA expression analysis showed that microRNA changes differ between male and female P301S mice, differently influencing tau pathology in each sex (Kodama et al., 2020). In this study, depletion of Dicer, an RNase III endonuclease (Song & Rossi, 2017) was used to determine the accumulation of miRNA precursors in P301S mice. Tau inclusions were more frequent in male P301S/Dicer KO mice than in their female counterparts in the cortex, amygdala, and piriform cortex. Consistently, male P301S/Dicer KO mice displayed more amoeboid-like microglia. Microglia bulk sequencing revealed that male microglia were enriched with genes involved in inflammation and phagocytosis, including Spp1, Ccl6, Lpl, Il1b, and Cst7. Single cell sequencing confirmed that sex-dependent microRNAs cluster enrichment supports the differential microglial response to tau pathology, as male P301S/Dicer KO exhibited increased disease-associated signatures and decreased homeostatic microglia compared to females. Conversely, in the rTg4510 mouse, expressing a repressible P301L tau transgene, female mice have been reported to show more severe cognitive impairment and higher levels of phosphorylated tau than male mice (Yue, Hanna, Wilson, Roder, & Janus, 2011). Sex differences have also been observed in the JNPL3 strain, which expresses mutant P301L tau driven by the mouse prion promoter, with higher numbers of NFTs and tau expression levels observed in female mice compared with males (Lewis et al., 2001). Whether different tau mutations used to generate these lines contribute to sex differences in tau-induced deficits remains to be determined.

Nontransgenic animals

To our knowledge, very few studies have used both sexes with nontransgenic animals, and most have used intracerebroventricular (ICV) injections of streptozotocin (STZ) to model AD-like phenotypes in rats and mice. Biasibetti and colleagues demonstrated that the behavioral effects and changes in neurochemical markers depended on sex and were more prominent in males (Biasibetti et al., 2017). In particular, in the object recognition test, a nonspatial memory test, a significant cognitive impairment was present in the male groups treated with STZ, whereas female rats were more resistant to STZ effects. Male STZ mice also displayed reduced choline acetyltransferase content, while in females, a significant decrease was present only 8 weeks after STZ infusion. Finally, in this model, male STZ mice showed reduced glucose metabolism, more activated hippocampal astrocytes, and oxidative imbalance in the hippocampus. Similarly, Bao and colleagues showed that females were more resistant to the learning and memory impairment induced by STZ administration (Bao et al., 2017). Moreover, male STZ mice had more tau hyperphosphorylation and Aβ40/42, and increased GSK-3β and BACE1 activities, with more pronounced loss of dendritic and synaptic plasticity.

Limitations

While AD is a strictly human disease, the generation of genetically modified mice expressing mutations in genes that cause AD has enabled progress in understanding its pathogenesis.

Most AD model studies show that females have more severe disease pathology and progression compared to males, which aligns with observations in humans. However, many factors should be considered when drawing conclusions on how sex influences AD pathology in preclinical models. As reviewed here, different mutations could worsen AD pathology in either male or female mice. Moreover, the type of promoter used might also have an impact on the outcome of behavioral and neuropathological tests. Yang et al., in 2006, reported a large-scale analysis of whole brains from more than 150 male and female mice, highlighting the degree of sex-dependent gene expression in the mammalian brain (Yang et al., 2006). Their study showed that gene expression levels in 14% of genes expressed in the brain were influenced by sex, with most of these genes being located on autosomal chromosomes. Such differences would be traditionally attributed to hormonal regulation, but in more recent studies, genetic and epigenetic effects have been associated with the inheritance of the X and Y chromosomes (Ratnu, Emami, & Bredy, 2017). Thus, sex-specific differences in epigenetic regulation may also influence promoters’ expression, and hence mouse behavior and transgene pathology. Examples of this are the promoter for brain-derived neurotrophic factor (BDNF), which has been reported to be hypermethylated in female mice (Baker-Andresen, Flavell, Li, & Bredy, 2013), or the ERαpromoter, more methylated in male mice in the amygdala (Edelmann & Auger, 2011).

Taken together, the evidence suggests that different promoters may possess different activities in driving transgene expression in mice of different sexes. Moreover, what appears more challenging would be imitating the human reproductive decline in mouse models of AD, as mouse aging is not characterized by lowered gonadal steroid levels (Nelson, Felicio, Osterburg, & Finch, 1992; Nelson, Latham, & Finch, 1975), as occurs in men and, in particular, during menopause in women, during aging (Chakravarti et al., 1976; Ferrini & Barrett-Connor, 1998) (see Chapter 9). The other translation obstacle is represented by the difference in reproductive senescence, which is due to dysregulation of the neuroendocrine system in female rodents (Kermath & Gore, 2012), while in women, there is a decline of oocyte numbers and ovary function (Fitzgerald, Zimon, & Jones, 1998). Appropriate modeling could be accomplished by gonadectomy in rodents, recapitulating a critical aspect of human reproductive aging in males and females (Carroll et al., 2010). However, mouse studies on hormonal effects in AD have shown that gonadal steroids benefit cognition in the normal aging brain in both female and male mice (Dubal, Broestl, & Worden, 2012), raising several important questions about how to investigate the loss and replacement of hormones in models of AD.

Conclusion

Transgenic manipulation and mutations should be considered when interpreting and translating results from animal models to humans. Given that animal studies have helped to understand sex-dependent aspects in AD, sex differences should always be taken into account when examining the pathological characteristics of AD and related target molecules (Fig. 1). One should take sex differences in animals into account in both mechanism research and drug design—for example, the emerging role of novel sex-specific EEG fingerprints as potential early biomarkers of AD in the future, or highlighting the importance of considering sexes when evaluating the effect of long-term antibiotic treatment on the microbiome and peripheral inflammation. Moreover, treatment with VPA might not show the same beneficial results in male and female AD mice, as their sex may significantly influence AD. Thus, VPA may be a promising remedy for AD only if basic biological differences and sex specificity are taken into account. To develop potential therapeutic targets in postmenopausal female ApoE4 carriers, future studies with ovarian hormone manipulation in ApoE4 female rodents are necessary to probe the mechanisms underlying AD pathology. Moreover, further animal investigation will also be required to define the mechanisms driving APOE and sex differences in microglial function. Considering the sex differences existing in virtually all AD animal models in use (see Table 1), single sex-studies should be a thing of the