Neurophysiological Assessment of Brain Network Activity ...

FACULTY OF PSYCHOLOGY AND EDUCATIONAL SCIENCES

Neurophysiological Assessment of

Brain Network Activity Involved in

Cognitive Processing in Animal

Models of Alzheimer’s Disease

Sofia Jacob

Doctoral thesis offered to obtain the degree of Doctor in Philosophy (Ph.D.)

Department of Brain and Cognition

Supervisor (s): Prof. Dr. Detlef Balschun Prof. Dr. Wilhelmus H.I.M. (Pim) Drinkenburg

2020

Neurophysiological Assessment of Brain

Network Activity Involved in Cognitive

Processing in Animal Models of Alzheimer’s

Disease

Sofia Jacob

Promoter (s):

Prof. Dr. Detlef Balschun

Prof. Dr. Wilhelmus H.I.M. Drinkenburg

Jury Members:

Dr. Per Nilsson

Prof. Dr. Jos Prickaerts

Prof. Dr. Cees Van Leeuwen

Prof. Dr. Andreas Van Leupoldt

Leuven, 3rd April 2020

Thesis submitted in partial fulfillment of the requirements for the degree of Doctor in Philosophy

at KU Leuven, Belgium

Department of Brain and Cognition

Faculty of Psychology & Educational Sciences

Katholieke Universiteit Leuven

“In god we trust, all others must bring data”

W. Edwards Deming

P a g e 1 | 128

Summary

The pathophysiological processes of Alzheimer’s disease (AD) are thought to start 20 years before

cognitive symptoms are observed for the first time. During the initial stage, known as the pre-

symptomatic phase of AD, also depicted as the preclinical phase, small alterations in the brain,

unnoticeable to the affected individual, start to occur. With disease progression, these small

changes advance into irreversible brain damage. It is believed that the best chance of therapeutic

success in AD will be early intervention. Biomarker research has become one of the main

investigational areas of AD as they could play an instrumental role in unequivocally identifying

the initial phase of AD.

Accumulating evidence suggests that neuronal oscillations play an important role in driving

brain network communications. Furthermore, oscillatory alterations are commonly observed in

patients with AD. It is still unclear whether they are early driving mechanisms of cognitive

dysfunction. If these neuronal network alterations can be identified at the preclinical phase of AD,

they could be implemented as a disease diagnostic tool.

Numerous animal models recapitulating the hallmarks of AD pathogenesis have been

created to facilitate the understanding of the molecular mechanisms underlying the disease process.

Among the most common models are transgenic animals that produce amyloid-β (Aβ) pathology

due to the artificial overexpression of the human amyloid precursor protein (APP) with mutations

linked to familial AD. More recent models include the App knock-in mice that produce robust Aβ

amyloidosis with physiological App expression levels.

The primary goal of this study was to investigate electrophysiological readouts in

combination with cognitive tasks to characterize electrophysiological functional alterations at ages

relevant for the preclinical AD phase. Two animal models were used, one that overexpresses

mutated human APP (the McGill-R-Thy1-APP rat) and another one that expresses mutated

humanized App at physiological levels (AppNL-G-F mice). We hypothesized that in these models, at

an age that mimics the preclinical stage of AD, Aβ amyloidosis causes aberrant network activity,

which reflects the early development of cognitive disturbances. Our results from the AppNL-G-F

characterization study do not support the hypothesis of early alterations in cognition relevant

oscillations due to Aβ amyloidosis. This study also indicated that APP overexpression, and not Aβ

overproduction, might be responsible for the abnormal network activity in the McGill-R-Thy1-

APP rat model. More in general, the experimental approach presented in this thesis provides a

P a g e 2 | 128

versatile methodology for assessment of complex neuronal network dynamics in models of AD as

well as in models of other neurodegenerative diseases.

P a g e 3 | 128

Samenvatting

Het begin van de pathofysiologische processen die aanleiding geven tot de ziekte van Alzheimer

wordt aangenomen zich te situeren 20 jaar voordat cognitieve symptomen voor het eerst worden

waargenomen. Tijdens de initiële fase van de ziekte, bekend als de presymptomatische ofwel

preklinische fase, manifesteren zich kleine veranderingen in de hersenen, onmerkbaar voor het

getroffen individu. Naarmate de ziekte vordert, evolueren deze kleine veranderingen tot

onomkeerbare hersenschade. Men neemt aan dat vroege interventie de grootste kans op

therapeutisch succes zal hebben. Zodoende is het onderzoek naar biomerkers één van de

belangrijkste onderzoeksgebieden van ziekte van Alzheimer, aangezien deze een instrumentele rol

kunnen spelen bij het accuraat identificeren van de preklinische fase.

Er is steeds meer bewijs dat gesynchroniseerde electrische activiteit van hersencellen, de

zogenaamde neuronale oscillaties, een belangrijke rol spelen in de communicatie van neuronale

netwerken in het brein. Bovendien worden afwijkingen in bepaalde oscillaties vaak waargenomen

bij Alzheimerpatiënten. Het is nog onduidelijk of dit vroege mechanismen zijn, die gerelateerd

zijn aan cognitieve disfunctie. Als dergelijke neuronale netwerkveranderingen in de preklinische

fase van de ziekte van Alzheimer kunnen worden geïdentificeerd, dan kunnen ze worden

geïmplementeerd als diagnostisch hulpmiddel.

Talrijke diermodellen die de hoofdkenmerken van de Alzheimer pathogenese recapituleren

werden gecreëerd om inzichten te genereren in de moleculaire mechanismen, die ten grondslag

liggen aan het ziekteproces. De meest voorkomende modellen zijn transgene dieren die amyloïd-

β (Aβ) pathologie produceren als gevolg van de kunstmatige overexpressie van het menselijke

amyloïde precursor eiwit (Engelse afkorting: APP), met mutaties gelinkt aan familiale vormen van

de ziekte van Alzheimer. Meer recente modellen omvatten de App knock-in muizen, die robuuste

Aβ amyloïdose vertonen, maar fysiologische App expressieniveaus hebben.

Het primaire doel van deze studies was om Alzheimer diermodellen te karakteriseren op

leeftijden die overeenkomen met de preklinische fase van Alzheimer. Dit werd uitgevoerd aan de

hand van technieken, die de elektrofysiologische functie meten in combinatie met cognitieve

taken. Twee diermodellen werden gebruikt, één die gemuteerd menselijk APP tot overexpressie

brengt (de McGill-R-Thy1-APP rat), en een andere die gemuteerd gehumaniseerd App op

fysiologische niveaus tot expressie brengt (de AppNL-G-F muizen). Onze hypothese was dat in deze

modellen, op een leeftijd die het preklinische stadium van de ziekte van Alzheimer simuleert, Aβ

P a g e 4 | 128

amyloïdose afwijkende netwerkactiviteit veroorzaakt, welke de vroege ontwikkeling van

cognitieve stoornissen representeert. Onze resultaten in de AppNL-G-F muizen ondersteunen de

hypothese van vroege veranderingen in voor cognitie relevante oscillaties als gevolg van Aβ

amyloïdose niet. Deze studie gaf verder aan dat APP overexpressie, en niet Aβ overproductie,

mogelijk verantwoordelijk is voor de abnormale netwerkactiviteit in het McGill-R-Thy1-APP rat

model. Meer in het algemeen biedt de experimentele benadering in dit proefschrift een veelzijdige

methodologie voor onderzoek van de complexe neuronale netwerkdynamiek in diermodellen van

de ziekte van Alzheimer, alsook diermodellen van andere neurodegeneratieve ziekten.

List of abbreviations

P a g e 5 | 128


A/T/N Amyloid deposition [A], neurofibrillary tangle [T] and neuronal injury [N]

AAALAC Accreditation of laboratory animal care international

AB Amyloid beta

ABCA7 ATP-binding cassette transporter A7

AD Alzheimer's disease

AICD amyloid precursor protein intracellular domain

ANOVA Analysis of variance

AP Anterior-posterior

APP Amyloid precursor protein

Cg Cingulate cortex

CLU Clusterin

CR1 complement receptor 1

CSF cerebrospinal fluid

CT Correction trial

CTF C-terminal fragment

dCA1 dorsal CA1 region of the hippocampus

ddPCR Droplet digital PCR

DMS dorsal medial striatum

DNA Deoxyribonucleic acid

DV dorsal-ventral

EEG Electroencephalography

ELISA enzyme-linked immunosorbent assay

fAD familial AD

FDR False discovery rate

Frt Ass Ctx frontal associated cortex

GABABR Amino aminobutyric acid type B receptor

GuHCl Guanidin-hydrochloride

HET Heterozygous

HFO high frequency oscillation

HFP high frequency oscillations


P a g e 6 | 128

HG High gamma

HO homozygous

ITI Inter-trial-interval

IWG International Working Group

KI Knock-in

Lat Ctx Lateral cortex

LFP Local field potentials

LG Low gamma

L-M1/M2 Ctx left M1/M2 cortex

LTP long term potentiation

M mean

MAPT microtubule-associated protein tau

MI Modulation index

ML Medial-lateral

mPFC medial prefrontal cortex

NFTs neurofibrillary tangles

NIA/AA National Institute on Aging/Alzheimer Association

PAC phase-amplitude coupling

PBS Phosphate-buffered saline

PCR Polymerase chain reaction

PDGF factor B-chain

PET positron emission tomography

PrP prior protein

PS Presenilin

PSD Power Spectral Density

PSEN1 presenilin 1

PSEN2 presenilin 2

R-M1/M2 Ctx right M1/M2 cortex

ROI Regions-of-interest

RSC retrosplenial cortex

s Soluble


P a g e 7 | 128

S- Unconditioned stimulus

S+ Conditioned stimulus

sAD sporadic AD

SD Standard deviation

Thy-1 thymocyte differentiation antigen 1

TREM2 triggering receptor expressed on myeloid cells 2

TUNL Trial-unique delayed non-matching-to-location

V1 Ctx Left V1 cortex

VD Visual Discrimination

WT Wildtype

P a g e 8 | 128

Table of Contents

P a g e 9 | 128

Table of Contents

General Introduction .................................................................................................................. 13

1.1. Introduction to Alzheimer’s Disease ................................................................................................ 13

1.2. Etiology of Alzheimer’s Disease ...................................................................................................... 13

Familial Alzheimer’s Disease .............................................................................................................. 14

Sporadic Alzheimer’s Disease ............................................................................................................. 14

1.3. Pathological Hallmarks ..................................................................................................................... 14

APP Processing and the Role of Aβ Pathology ................................................................................... 16

Neurofibrillary Tangles ....................................................................................................................... 18

Synaptic and Neuronal Loss ................................................................................................................ 18

1.4. Alzheimer’s Disease Progression ..................................................................................................... 19

1.5. Oscillations ....................................................................................................................................... 20

Oscillations in Pathological Conditions............................................................................................... 21

1.6. Animals Models of Alzheimer’s Disease Pathology ........................................................................ 22

Objectives ................................................................................................................................... 27

2.1. General Objective of the Project ....................................................................................................... 27

2.2. Specific Objectives ........................................................................................................................... 27

Objective 1: .......................................................................................................................................... 27

Objective 2: .......................................................................................................................................... 27

Objective 3: .......................................................................................................................................... 27

Material and Methods ................................................................................................................ 29

3.1. Animals ............................................................................................................................................. 29

McGill-R-Thy1-APP Rats ................................................................................................................... 29

AppNL-G-F Mice ..................................................................................................................................... 29

3.2. Surgery .............................................................................................................................................. 30

3.3. Behavioral Task ................................................................................................................................ 31

Trial-unique delayed Non-matching-to-location Task ........................................................................ 31

Apparatus ......................................................................................................................................... 32

Touchscreen Pre-training Stages ..................................................................................................... 33

TUNL Task Acquisition .................................................................................................................. 33

TUNL Task Delay Test ................................................................................................................... 34

Visual Discrimination Task ................................................................................................................. 35

Apparatus ......................................................................................................................................... 36

Touchscreen Pre-training Stages ..................................................................................................... 36

VD Task Tests ................................................................................................................................. 37

Table of Contents

P a g e 10 | 128

3.4. Electrophysiological Measurements ................................................................................................. 39

Recordings ........................................................................................................................................... 39

Home-cage Exploration: .................................................................................................................. 39

TUNL Task: ..................................................................................................................................... 40

VD Task: .......................................................................................................................................... 40

Relative and Absolute Power Spectral Density Analysis .................................................................... 40

Phase-amplitude Coupling Analysis .................................................................................................... 40

3.5. Brain Aβ1-42 Enzyme-linked Immunoassay (ELISA)..................................................................... 41

3.6. Immunohistochemistry ..................................................................................................................... 42

3.7. McSA1 Antibody Specificity ........................................................................................................... 43

3.8. Genotyping ....................................................................................................................................... 44

Droplet Digital PCR (ddPCR) ............................................................................................................. 44

3.9. Statistical Analysis............................................................................................................................ 44

Behavioral, Electrophysiological and Histopathological Characterization of McGill-R-Thy1-

APP Rats ...................................................................................................................................................... 47

4.1. Introduction....................................................................................................................................... 47

4.2. Results............................................................................................................................................... 48

Performance during TUNL Task ......................................................................................................... 48

Pre-training: ..................................................................................................................................... 48

TUNL Task Acquisition: ................................................................................................................. 49

TUNL Task Delay Test ................................................................................................................... 50

Brain Oscillations Analysis During TUNL Task ................................................................................. 52

Power Spectral Density .................................................................................................................... 52

Phase-amplitude Coupling ............................................................................................................... 57

Brain Oscillation Analysis During Home-Cage Environment Exploration......................................... 61

Power Spectral Density .................................................................................................................... 61

Phase-Amplitude Coupling .............................................................................................................. 62

Pathology ............................................................................................................................................. 63

Immunohistochemistry .................................................................................................................... 63

McSA1 Antibody Specificity Analysis ........................................................................................... 67

Genotyping – ddPCR ........................................................................................................................... 68

4.3. Conclusion ........................................................................................................................................ 68

Behavioral, Electrophysiological and Histopathological Characterization of AppNL-G-F Mice .. 71

5.1. Introduction....................................................................................................................................... 71

5.2. Results............................................................................................................................................... 73

Performance during VD Task .............................................................................................................. 73

Pre-training: ..................................................................................................................................... 73

Table of Contents

P a g e 11 | 128

Visual Discrimination Task: ............................................................................................................ 73

Brain Oscillations Analysis During VD Task ..................................................................................... 75

Relative Power Spectral Density: .................................................................................................... 75

Phase-Amplitude Coupling:............................................................................................................. 77

Brain Oscillations Analysis During Home-cage Environment Exploration ........................................ 83

Relative Power Spectral Density: .................................................................................................... 83

Phase-amplitude Coupling: .............................................................................................................. 84

Pathology ............................................................................................................................................. 85

5.3. Conclusion ........................................................................................................................................ 87

General Discussion .................................................................................................................... 89

References .................................................................................................................................................... 95

Scientific Contributions ............................................................................................................................. 111

Curriculum Vitae ....................................................................................................................................... 113

Acknowledgements .................................................................................................................................... 117

P a g e 12 | 128

P a g e 13 | 128

General Introduction

1.1. Introduction to Alzheimer’s Disease

Alzheimer’s disease (AD) is the most common cause of progressive neurodegenerative dementia

(Scheltens et al., 2016). The characteristic symptoms are impairments in memory, language,

problem-solving and other cognitive skills. In the more advanced stages, people with AD need

around-the-clock care, making it a devastating disease, not only for the patients, but also for their

caregivers. Considering that the main risk factor of the disease is age (Querfurth & LaFerla, 2011)

and that average life-expectancy increases, the associated personal, social and socio-economic

burden will intensify, if we do not find an effective treatment. The major pathological hallmarks

of AD are the formation and deposition of amyloid beta (Aβ) senile plaques and neurofibrillary

tangles (NFTs) (Scheltens et al., 2016). These pathological changes are accompanied by

neuroinflammation, aberrant synaptic and neuronal network activities (Başar et al., 2016;

Friedman, Honig, & Scarmeas, 2013; Nimmrich, Draguhn, & Axmacher, 2015), and eventually

dramatic brain shrinkage due to neuronal damage (Scheltens et al., 2016).

Despite the advances in AD research, much is still to be discovered about the etiology of

the disease and how it can be prevented, slowed or stopped. Current treatments are purely

symptomatic with no disease modifying effects, providing only temporary improvements. There

is an urgent need for early diagnosis to avoid irreversible brain damage and to allow timely

intervention of potential disease modifying drugs.

1.2. Etiology of Alzheimer’s Disease

In the majority of AD patients, the etiology of the disease is unknown and it develops as a result

of multiple factors. This condition is described as sporadic AD (sAD). In no more than 1 percent

of cases, the disease is caused by autosomal dominant inherited mutations and are defined as

familial AD (fAD). The majority of patients develop AD at age 65 or older. These cases are called

late-onset AD and the etiology is always sporadic. Early-onset develops before age 65 and it can

be either sAD or fAD (Scheltens et al., 2016).

P a g e 14 | 128

Familial Alzheimer’s Disease

Mutations in amyloid precursor protein (APP), presenilin 1 (PSEN1), presenilin 2 (PSEN2) genes

are the only three associated with fAD. People inhering APP or PSEN1 mutations are guaranteed

to develop AD, while those who inherited a PSEN2 mutation have a 95 percent chance to develop

the disease (Bekris, Yu, Bird, & Tsuang, 2010). Some mutations in these three genes affect APP

cleavage and Aβ production. Importantly, despite the low incidence of fAD, multiple transgenic

animal models carrying mutations in these genes have been created to further understand the

molecular mechanisms implicated in the pathogenesis of the disease (see section –Animals Models

of Alzheimer’s Disease Pathology).

Sporadic Alzheimer’s Disease

Most AD cases develop as a result of multiple environmental and genetic factors rather than a

single cause, but unfortunately little is known about the interplay between them (Alzheimer’s

Association, 2019; Livingston et al., 2017). From the environmental risk factors, age is the

greatest. Modifiable risk factors include cardiovascular disease risk factors, such as smoking, and

diabetes, lower education, lack of exercise, and traumatic brain injury (Alzheimer’s Association,

2019). Genome-wide association studies have identified single nucleotide polymorphisms in

several genes that are associated with an increased risk factor, the most influential being the APOE

ε4 allele. APOE plays an important role in cholesterol transport and Aβ clearance. There are three

common alleles in the APOE gene (ε2, ε3, and ε4). Having the ε4 form increases the risk of getting

sAD compared to the ε3 form, and having the ε2 form may decrease the risk compared with having

ε3 form (Alzheimer’s Association, 2019; Spinney, 2014). Other single nucleotide polymorphisms

have been identified, for example, in Clusterin (CLU), ATP-binding cassette transporter A7

(ABCA7), triggering receptor expressed on myeloid cells 2 (TREM2), complement receptor 1

(CR1), and CD33 genes as genetic risk factors of sAD (Karch & Goate, 2015).

1.3. Pathological Hallmarks

In 1907 Alois Alzheimer identified the two main pathological hallmarks of AD: Extracellular

amyloid plaques and intracellular NFTs (Stelzmann, Norman Schnitzlein, & Reed Murtagh, 1995).

Eighty years after this discovery, the molecular compositions of these pathological hallmarks were

P a g e 15 | 128

described (Glenner & Wong, 1984; Grundke-Iqbal et al., 1986). NFTs are composed of aggregated

hyper-phosphorylated forms of the microtubule-associated protein Tau (Grundke-Iqbal et al.,

1986), while plaques are mainly composed of insoluble Aβ aggregates (Glenner & Wong, 1984).

Macroscopically, AD is characterized by general cortical mass reduction, enlargement of

ventricles, and hippocampal and cerebral cortex atrophy (Apostolova, 2013). See Figure 1.

Figure 1. Pathological hallmarks of Alzheimer’s disease (AD). A) atrophy of the brain. Depicts a

section of a hemibrain from an AD patient (left) and a healthy aged brain (right). The brain from

the AD patient shows marked atrophy, dilation of the lateral ventricle, and a small hippocampus.

B) Neuritic plaques [P] and neurofibrillary tangles [N] in the hippocampus, as seen with the

modified Bielschowshy silver impregnation. C) Immunostaining of a frontal cortex section with

an anti-amyloid-β antibody (10D5) revealing a diffuse amyloid plaque [D], a dense cored plaque

[C] and cerebral amyloid angiopathy [A]. D) immunodetection of neurofibrillary tangles (N) and

neuritic plaques [P] in frontal lobe. Figure adapted from (Wippold, Cairns, Vo, Holtzman, &

Morris, 2008).

P a g e 16 | 128

APP Processing and the Role of Aβ Pathology

Aβ peptides are generated by sequential proteolytic cleavage of APP. Depending on the secretases

that cleave APP, it can undergo amyloidogenic and non-amyloidogenic processing. In the non-

amyloidogenic pathway, APP is first cleaved by α secretase in the middle of the Aβ sequence to

generate the soluble APPα (sAPPα) fragment and APP-CTF-α. The subsequent cleavage of APP-

CTF-α by γ-secretase yields the P3 peptide and the amyloid precursor protein intracellular domain

(AICD). In the amyloidogenic pathway, APP is cleaved by β-secretase to produce sAPPβ and a C-

terminal fragment (APP-CTF-β) which contains the Aβ peptide. In turn, APP-CTF-β is cleaved by

γ-secretase at multiple sites in the plasma membrane yielding Aβ peptides of different lengths and

AICD (see Figure 2, right panel) (Citron, 2004; Maya & Bassem, 2014). Aβ protein fragments can

accumulate and build up into plaques. These plaques can exhibit different sizes and degree of

compactness. For instance, diffuse Aβ plaques are mostly present in nondemented elderly people,

while neuritic Aβ plaques are commonly found in AD patients. From the different Aβ peptides,

the Aβ1-42 is considered the main constituent of the amyloid plaques (Serrano-Pozo, Frosch,

Masliah, & Hyman, 2011). Although cleavage of APP into Aβ had been extensively studied in

relation to AD, the relationship of physiological α-secretase to β-secretase processing is not fully

understood.

The amyloid cascade hypothesis proposes Aβ accumulation and amyloid plaque deposition

as the first initiators of AD pathogenesis (Hardy & Higgins, 1992). The hypothesis was postulated

in the early 90’ and proposed a linear cascade where the deposition of Aβ is the initial pathological

event in AD, leading to amyloid plaques, NFTs, neuronal loss and ultimately to dementia (Hardy

& Higgins, 1992; Karran, Mercken, & De Strooper, 2011).

Several pieces of evidence made this hypothesis the leading one in the field guiding

academic and pharmaceutical research in the last two decades. First, all currently known mutations

associated with fAD affect Aβ production or aggregation. Mutations in APP, PSEN1, and PSEN2

(catalytic subunit of γ-secretase) affect APP cleavage and Aβ production, favouring the release of

longer and less soluble Aβ peptides (Bekris et al., 2010). Second, individuals with Down’s

syndrome are at increased risk of developing AD. Most of these patients have three copies of the

APP gene, leading to elevated APP expression and increased Aβ deposition (Lott, 2012). Third, a

coding mutation in the APP gene (A673T) has been showed to reduce the risk for AD (TCW &

Goate, 2017). In October 2019, Biogen announced the 2020 application for Food and Drug

P a g e 17 | 128

Administration approval for aducanumab, a human monoclonal antibody against Aβ developed.

The results of the phase 3 clinical study are the first ones to demonstrate that clearance of Aβ

aggregates can reduce clinical decline in patients with early AD

(https://www.alzforum.org/news/research-news/reports-my-death-are-greatly-exaggerated-

signed-aducanumab). If this drug works and it is approved, it will become the first treatment for

AD.

Despite the above supporting evidence, the linearity of the amyloid hypothesis remains one

of the most highly debated topics in the field. Some of the scepticism on this hypothesis comes

from the lack of consistent relation between the extent of amyloid pathology and the severity of

dementia (Terry et al., 1991). The rate of cognitive impairment in AD correlates best with the

burden of NFT (Bierer et al., 1995; Pontecorvo et al., 2017). Furthermore, with the new

understanding of the preclinical AD phase (see section Alzheimer’s Disease Progression), a more

holistic approach where other cell populations such as microglia and astrocytes also play a role in

the progression of the diseases has been suggested (De Strooper & Karran, 2016).

Figure 2. The proteolytic processing of amyloid precursor protein (APP). APP can be cleaved by

α-, β- and γ-secretases; the cleavage sites of these proteases are indicated in the full-length APP

shown in the center of the figure. APP can undergo amyloidogenic (right) or non-amyloidogenic

(left) processing. In the amyloidogenic pathway, cleavage by β-secretase results in the formation

of soluble APPβ (sAPPβ) and APP-CTF-β. The subsequent action of γ-secretase on APP-CTF-β

releases Aβ from the amyloid precursor protein intracellular domain (AICD). In the non-

amyloidogenic pathway, cleavage by α-secretase prevents the formation of Aβ; α-secretase cleaves

within the Aβ sequence, giving rise to sAPPα and the membrane-tethered APP-CT-α, which in

https://www.alzforum.org/news/research-news/reports-my-death-are-greatly-exaggerated-signed-aducanumab

https://www.alzforum.org/news/research-news/reports-my-death-are-greatly-exaggerated-signed-aducanumab

P a g e 18 | 128

turn is cleaved by γ-secretase resulting in release of the P3 peptide and AICD. Figure adapted from

(Maya & Bassem, 2014).

Neurofibrillary Tangles

Tau, the main component of NFTs is a microtubule-associated protein, essential for axonal

microtubule assembly and stability (Lindwall & Cole, 1984). When hyper-phosphorylated, tau

loses affinity for microtubules becoming highly prone to aggregation, and filament formation

(Abraha et al., 2000). NFTs pathology in AD develops in a highly distinct spatial manner. The first

tau depositions are observed in the trans-entorhinal region, followed by the hippocampal

formation, and finally the rest of the neocortex (Braak & Braak, 1991).

It is important to notice that NFTs are common in many neurodegenerative diseases

without Aβ plaques, such as frontotemporal dementia and Pick’s disease. Together, these diseases,

are known as tauopathies (Spillantini & Goedert, 2013). In the context of AD, a substantial amount

of evidence suggests an interplay between Aβ and tau (Ittner & Götz, 2011).

Synaptic and Neuronal Loss

A major pathological characteristic found in AD patients is synaptic degeneration (Selkoe, 2002).

Importantly, the degree of cognitive decline in patients with AD has been more robustly correlated

to synaptic loss than with the number of NFTs and Aβ plaques (DeKosky & Scheff, 1990; Terry

et al., 1991). Furthermore, synaptic loss in areas associated with memory, such as the

hippocampus, has been reported in the early stages of AD (Scheff, Price, Schmitt, & Mufson,

2006).

The interaction between Aβ and tau and their role on the cascades that lead to synaptic

dysfunction remains a controversial issue and a topic of active investigation. For instance, it has

been suggested that Aβ modulates synaptic transmission (Parihar & Brewer, 2010). The Aβ

oligomer hypothesis places these soluble Aβ aggregates as the key neurotoxic agents that cause

synaptic failure (Karran & De Strooper, 2016). Tau aggregation may play a role in synaptic

dysfunction by causing axonal transport deficits (Mandelkow, Stamer, Vogel, & Thies, 2003).

While new evidence suggest that intermediate forms of tau, prior to NFT formation, are the toxic

species (Shafiei, Guerrero-Muñoz, & Castillo-Carranza, 2017). Furthermore, microglial activation

and neuroinflammation can affect synaptic activity (Karran & De Strooper, 2016).

P a g e 19 | 128

1.4. Alzheimer’s Disease Progression

In the last decades, a new understanding of AD progression has developed. Currently, three stages

of AD are identified: preclinical AD, mild cognitive impairment (MCI) due to AD, and dementia

due to AD (see Figure 3) (Sperling et al., 2011). The initial preclinical stage starts around 20 years

before cognitive decline. This stage is characterized by biomarker changes without any overt

neurological symptoms. During MCI due to AD, some cognitive symptoms on top of the

biomarker readouts start to emerge. At the last stage, several symptoms of dementia from mild to

moderate to severe are accompanied by irreversible neurodegeneration (Jack et al., 2010; Sasaguri

et al., 2017).

Figure 3. Model of Alzheimer’s disease (AD) trajectory. The stage of preclinical AD precedes

mild cognitive impairment (MCI) and encompasses the spectrum of pre-symptomatic autosomal

dominant mutation carriers, asymptomatic biomarker-positive older individuals at risk for

progression to MCI due to AD and AD dementia, as well as biomarker-positive individuals who

have demonstrated subtle decline from their own baseline that exceeds that expected in typical

aging, but would not yet meet criteria for MCI. Note that this diagram represents a hypothetical

model for the pathological-clinical continuum of AD but does not imply that all individuals with

biomarker evidence of ongoing AD-pathophysiology will progress to the clinical phases of the

illness. Figure adapted from (Sperling et al., 2011).

Classification criteria based on biomarkers from the International Working Group (IWG),

the National Institute on Aging/Alzheimer Association (NIA/AA) (Visser, Vos, van Rossum, &

Scheltens, 2012), and more recently the A/T/N (amyloid deposition [A], neurofibrillary tangle [T]

and neuronal injury [N]) classification had been proposed (Dubois et al., 2016; Jack, Hampel,

Universities, Cu, & Petersen, 2016). In the A/T/N criterion “A” refers to the value of an Aβ

P a g e 20 | 128

biomarker (amyloid positron emission tomography [PET] or cerebrospinal fluid [CSF] Aβ42); “T”

refers to the value of a tau biomarker (CSF phosphor tau, or tau PET), and “N” biomarkers of

neurodegeneration or neuronal injury ([(18)F]-fluorodeoxyglucose-PET, structural MRI, or CSF

total tau). Each biomarker category is rated as positive or negative. An individual score might

appear as A+/T+/N-, or A+/T-/N-, etc. Importantly, these criteria do not specify disease labels and

thus do not represent a diagnostics classification system (Jack et al., 2016).

Biomarkers, such as PET scans and CSF tests, have tremendously advanced AD research

by providing relevant AD-related pathophysiology in living persons (Dubois et al., 2016; Jack et

al., 2016; Visser et al., 2012). It is believed that earlier interventions at the preclinical stage of AD

may offer the best chance of therapeutic success and biomarkers play an instrumental role in

correctly identifying the initial stage of AD. Biomarkers are not only being used as part of

diagnostic tools, but they also facilitate clinical trials targeting early stages of the disease, help to

monitor treatment responses, and can advance our knowledge on preclinical AD.

The search and investigation of more reliable, affordable, specific, sensitive, and less

invasive novel biomarkers to complement and replace the currently used markers, which are

expensive and invasive, is one of the main research areas in AD in these days, as targeted by

initiatives like the Alzheimer’s Drug Discovery Foundation Diagnostics Accelerator Initiative

(https://www.alzdiscovery.org).

1.5. Oscillations

Cognitive processes are based on the coordinated interaction of populations of neurons that are

distributed within and across different brain areas. Neuronal oscillations, transient and rhythmic

variation in neuronal activity, are usually assessed at frequencies from <0.1 to 300 Hz and

classified in frequency bands (delta, theta, alpha, beta, gamma, and high frequency oscillation

[HFO]). These bands are created based on correlations with response to cognitive processes,

vigilance state, pharmacology, and sensory and motor events, as different frequencies provide

distinct temporal windows for processing (Buzsáki & Draguhn, 2004; Jobert et al., 2012).

Accumulating evidence suggests that neuronal oscillations play a substantial role in driving the

communication in brain networks, not only in humans but also in rats (Fries, 2015; Hasselmo &

Stern, 2014; Herrmann, Strüber, Helfrich, & Engel, 2016).

https://www.alzdiscovery.org/

P a g e 21 | 128

Oscillation in the theta (4 -12 Hz) and gamma (30 - 100 Hz) band are commonly related to

cognitive processes and the changes in these oscillatory activities are correlated with specific task

demands (M. X. Cohen, 2014). Complex interactions between oscillations have also been

implicated in driving brain network’s communication. In particular, phase-amplitude coupling

(PAC) between the phase of theta and the amplitude of gamma has been suggested to be a potential

mechanism to regulate neuronal communication in multiple brain regions (Axmacher et al., 2010;

Lisman & Jensen, 2013; Tort et al., 2008). For instance, Tort and colleagues, using an item-context

association learning paradigm, investigated theta-gamma PAC in the CA3 region of the

hippocampus while rats learned the task. They observed that an increase in theta-gamma PAC was

correlated with an increase in accuracy during learning. Furthermore, the increased PAC was

maintained during over-trained sessions (Tort, Komorowski, Manns, Kopell, & Eichenbaum,

2009). Similarly, in humans, using transcranial alternating current stimulation in the medial

prefrontal cortex (mPFC) during a working memory task, Alekseichuk and coworkers

demonstrated that increased performance coincided with gamma bursts being phase-locked to the

peak of the theta oscillation (Alekseichuk, Turi, Amador de Lara, Antal, & Paulus, 2016).

Oscillations in Pathological Conditions

A better understanding of neuronal network interactions and circuit dynamics is crucial for our

comprehension of brain disorders. Electroencephalography (EEG) is a particularly attractive tool

to study alterations in circuit dynamics in AD and other neurodegenerative disorders due to its low

cost and easy implementation (Babiloni et al., 2020; Walsh, Drinkenburg, & Ahnaou, 2017).

In AD patients, changes in EEG spectral power or synchronization have been commonly

reported (Coben, Danziger, & Storandt, 1985; Engels et al., 2015; Nimmrich et al., 2015;

Voevodskaya et al., 2018). These oscillatory alterations are probably related to multiple direct and

indirect effects of Aβ and tau pathology leading to synaptic dysfunctions that impact local network

activity (Babiloni et al., 2020). A shift in power from fast to slow waves is typically reported in

EEG recordings with progressing AD pathology (Başar et al., 2016). For instance, decrease of

alpha power is commonly observed in the EEG of patients with dementia due to AD. Engels et al.

investigated functional connectivity in relation to disease severity. They divided AD patients in

mild, moderate and severe groups based on the Mini Mental State Examination and recorded EEG

to then perform functional connectivity analysis. Their results indicate a decrease in functional

P a g e 22 | 128

connectivity in the lower alpha band as the disease progresses (Engels et al., 2015). PAC has also

been shown to be affected in patients with AD (Poza et al., 2017) and MCI (Dimitriadis, Laskaris,

Bitzidou, Tarnanas, & Tsolaki, 2015). These disruptions are also present in animal models of

pathology (Busche & Konnerth, 2015; Palop et al., 2007; Palop & Mucke, 2010). Bazzigaluppi

and colleagues, recently demonstrated reduced PAC in the hippocampus and mPFC in a transgenic

rat model of AD at early stages of pathology (Bazzigaluppi et al., 2018).

Although some of the studies investigating oscillatory changes in patients used new

guidelines for diagnosis and investigated preclinical and clinical AD, the reliability of the AD

diagnosis in other studies must be considered with caution, given the lack of available biomarkers

at the time of the studies. For instance, it has been indicated a clinical misdiagnosis of about 20%

(Fischer et al., 2017).

Importantly, neuronal network changes are also seen in other pathologies such as

schizophrenia, bipolar disorders, and other neurodegenerative disorders. For instance, a disruption

of gamma oscillations, which in turn is associated with the cognitive deficits, has been observed

in these diseases (Nimmrich et al., 2015). Furthermore, it is still unclear whether neuronal

alterations occur rather at a late stage of the disease as a consequence of neurodegeneration, or at

an early stage as a primary mechanism contributing to cognitive dysfunction. If the second

conjecture is correct, the potential value of having a reliable electrophysiological biomarker of

preclinical and early stages of the AD could have a major impact on the disease diagnostics. Our

knowledge about the pathophysiology of neural networks in AD is limited and more research is

urgently needed.

1.6. Animals Models of Alzheimer’s Disease Pathology

Animal models have played a major role in AD research. To date, there are approximately 200

different transgenic rodent models (see https://www.alzforum.org/research-models). The most

common of them are transgenic mouse models overexpressing human APP, presenilin (PS), and/or

tau mutations. Even though none of the transgenic models replicates all aspects of AD, they have

provided important insights into the pathophysiology of the disease.

Transgenic mouse models overexpressing human APP were the first created models

(Games et al., 1995) and are still the most commonly used. These models overexpress APP with

https://www.alzforum.org/research-models

P a g e 23 | 128

one or multiple fAD mutations. Some of the most commonly used mutations are K670N/M671L

(Swedish) with increased total Aβ, V717F (Indiana), and V717I (London) with both having

increased Aβ42/ Aβ40 ratios (Goate et al., 1991; Karch & Goate, 2015; Murrell, Farlow, Ghetti,

& Benson, 1991). The APP is expressed from numerous promoters, such as factor B-chain

(PDGF), thymocyte differentiation antigen 1 (Thy-1), and prior protein (PrP) genes (Sasaguri et

al., 2017). Among the most frequently used rodent models are PSAPP, Tg2576, APP23, and J20.

These models are characterized by Aβ deposits, cognitive dysfunction, even before Aβ plaques,

and synaptic damages (Games et al., 1995; Hsiao et al., 1996; Mucke et al., 2000; Palop et al.,

2007; Sasaguri et al., 2017; Sturchler-Pierrat et al., 1997). At the brain network level, it has been

reported by numerous studies that these mouse models demonstrate various alterations (Busche &

Konnerth, 2015; Palop et al., 2007; Palop & Mucke, 2010), even before Aβ accumulation

(Goutagny, Gu, Cavanagh, Jackson, & Chabot, 2013). Although these results show some

phenotypical similarities with AD (Nimmrich et al., 2015; Palop & Mucke, 2009; Poza et al.,

2017), it is not clear what are the underlying mechanisms that cause the aberrant neuronal activity

observed in these models.

Other commonly used mouse models are double transgenics where APP mutant mice are

crossed with PSEN1 mutant mice. The most commonly used models carrying both mutant genes

are the APP/PS1 and 5xFAD (Holcomb et al., 1998; Oakley et al., 2006). These models are

characterized by rapid Aβ deposition, behavioral impairments, and neuronal loss. Single transgenic

mutant PSEN1 mice are not used as the overexpression of mutant PS1 alone does not induce AD

pathology or cognitive deficits (Hall & Roberson, 2012; Sasaguri et al., 2017). These models, as

well as the mutant APP models, fail to replicate two major pathologies of AD: NFT and substantial

neurodegeneration.

Several mutant tau transgenic mouse models presenting NFTs and neurodegeneration are

also available (Drummond & Wisniewski, 2017; Hall & Roberson, 2012; Kiyota, 2014). These

models express human tau with mutations that cause frontotemporal dementia with parkinsonism

linked to chromosome 17 (Yoshiyama, Lee, & Trojanowski, 2001). To investigate AD, better-

considered models are the ones combining tau transgenic models with APP models, such as the

APP-V717I x Tau-P301L mice (Terwel et al., 2008). These models exhibit both Aβ accumulation

and NFTs, making them more complete than other models and allowing to investigate interactions

between tau and Aβ pathology. However, the tau mutations used in these models are not directly

P a g e 24 | 128

linked to AD. Furthermore, the overexpression of APP and tau can have artificial phenotypes

(Born et al., 2014; Nilsson, Saito, & Saido, 2014).

Although most of the transgenic models are in mice, in the last decade, transgenic rat

models have also been created (Cohen et al., 2013; Leon et al., 2010). One of the main reasons for

this gap was related to the limited number of tools to alter the rat genome (Do Carmo & Cuello,

2013; Ellenbroek & Youn, 2016). However, rats have a rich and complex behavioral repertoire

and are physiologically, genetically, and morphologically closer to humans than mice (Do Carmo

& Cuello, 2013; Ellenbroek & Youn, 2016). Therefore, it is important to consider if this species is

a more relevant model to study AD pathology. One of the most characterized transgenic rat models

is the McGill-R-Thy1-APP, overexpressing human APP carrying both Swedish and the Indiana

mutations under the control of the murine Thy1.2 promoter (Leon et al., 2010). This model, similar

to the APP-overexpressing mouse models, develops Aβ deposits, cognitive impairments, as well

as synaptic and network decline (Parent et al., 2017; Qi et al., 2014).

All the above-mentioned transgenic rodent models have several important limitations.

First, these models use exogenous promoters to overexpress the different genes, which triggers an

abnormal expression of the proteins (Höfling et al., 2016). Second, the lack of standardization on

the use of different promoters makes it difficult to compare the models. Third, in the case of APP

overexpression models, various APP fragments, besides Aβ, are being produced (see Figure 2).

Therefore, while these models are useful to investigate Aβ production and deposition (Sasaguri et

al., 2017), the overexpression of these proteins makes it difficult to distinguish between the

phenotypes produced by Aβ pathology and other APP fragments unrelated to AD pathology.

In 2014, Saito and colleagues developed a new generation of App knock-in (KI) mice that

produce robust Aβ amyloidosis with physiological App protein levels (Saito et al., 2014). These

mice express humanized Aβ with either one (Swedish, AppNL), two (Swedish and

Beyreuther/Iberian, AppNL-F), or three (Swedish, Beyreuther/Iberian, and Artic, AppNL-G-F) fAD

mutations. The AppNL-F and AppNL-G-F models present Aβ plaques depositions and cognitive

alterations, while the AppNL do not develop plaques or have cognitive deficits (Masuda et al., 2016;

Saito et al., 2014a). More recently, Saito and colleagues developed another KI mouse model, in

this case with humanized microtubule-associated protein tau (MAPT) and crossed it with the App

KI models to investigate App-Tau interactions (Saito et al., 2019). Although, these models avoid

potential artifacts associated with the strong overexpression of AD-related genes, the development

P a g e 25 | 128

of pathology relies on several mutations linked to fAD and frontotemporal dementia and may not

accurately recreate the pathological processes in sAD.

Importantly, no existing model recapitulates all features of AD. Therefore, different models

may be most appropriate for addressing different questions and comparisons between different

models should always be done in the context of a specific scientific question rather than

considering just one particular model to be reliably mimicking AD (Jankowsky & Zheng, 2017).

P a g e 26 | 128

Objectives

P a g e 27 | 128

Objectives

2.1. General Objective of the Project

The general objective of this thesis was to characterize the dynamics in neural circuitry during

cognitive behavior in preclinical experimental rodent models of AD. We hypothesized that Aβ

amyloidosis causes aberrant network activity at the preclinical stage of the disease, which reflects

the early development of cognitive disturbances observed in models of AD pathology.

2.2. Specific Objectives

To evaluate this hypothesis, we have pursued the following major objectives:

Objective 1:

Study neuronal oscillations during cognitive process in a rat model of AD pathology at an age that

represents the preclinical phase of AD. For this purpose, we characterized the McGill-R-Thy1-

APP rat model at a behavioral, electrophysiological and histopathological level (chapter 4).

Objective 2:

Study neuronal oscillations during cognitive process in a mouse model of AD pathology at an age

that represents the preclinical phase of AD. For this purpose, we characterized the AppNL-G-F mouse

model at a behavioral, electrophysiological and histopathological level (chapter 5).

Objective 3:

Assess PAC as a sensitive functional readout of AD pathology. To this end, we investigated PAC

in relation to cognitive processes in the two AD models mentioned in objectives 1 and 2 (chapters

4 and 5).

Objectives

P a g e 28 | 128

To achieve the above-mentioned objectives the following techniques had to be established by

the candidate:

• Selection, assessment, and optimization of the trial-unique delayed non-matching-to-

location (TUNL) and visual discrimination (VD) tasks.

• Implementation, optimization, and validation of the wireless recording of local field

potentials (LFPs) during the behavioral task.

• Selection of electrophysiological readouts of interest.

• Implementation, optimization, and validation of surgical implantation of head-stages for

the recording of LFPs in rats and mice.

• Implementation, optimization, and validation of the synchronization between the video

system and the wireless electrophysiological recording to investigate oscillatory activity

during home-cage environment exploration.

• Detection and removal of artifacts from electrophysiological signal.

Overall, the approach presented in this thesis aimed at creating a versatile tool for further

assessment of the complex interplay between different frequency bands of oscillations of LFPs

and their relationship with behavior. Such a combination of techniques could open new avenues

for the investigation of cognition-related network perturbations in relevant animal models of AD

pathology, eventually determining their translational validity and potential use as

electrophysiological readouts in drug discovery and development. In this thesis we used this tool

in two transgenic model (objectives 1 and 2) and focus the oscillatory analysis in PAC (objective

3).

Material and Methods

P a g e 29 | 128


3.1. Animals

All experiments were performed in strict accordance with the guidelines of the Association for

Assessment and Accreditation of Laboratory Animal Care International (AAALAC) and with the

European Council Directive of 24 November 1986 (86/609/EEC) and European Ethics Committee

directive (2010/63/EU) for the protection of laboratory animals. In line with Belgian governmental

directives, all protocols were approved by the Animal Care and Use Committee of Janssen

Pharmaceutica NV. After weaning animals were singly housed (to prevent damage to chronically

instrumented head-stages by cage-mates) in individually ventilated cages under a reversed 12-12

light cycle (lights off 07:00-19:00; light intensity ~100 lux) under controlled environmental

conditions throughout the study (22 ± 2°C ambient temperature and relative humidity at 60%).

Home-cages were equipped with corn cob bedding, tissue for nesting material, a tinted

polycarbonate shelter for the mice and ad libitum water. Throughout pre-training and cognitive

testing, rodents were provided with a restricted diet (SAFE A05) to maintain them at 80% free-fed

weight to ensure consistent motivation towards reward pellets. All behavioral testing was

conducted during the dark phase to obtain optimal engagement in the behavioral task as this is the

active phase of the circadian cycle in mice and rats.

McGill-R-Thy1-APP Rats

In vivo data were obtained from 6 homozygous (HO), 6 heterozygous (HET), and 16 wildtype

(WT) male McGill-R-Thy1-APP rats (generated by Cuello and colleagues (Leon et al., 2010) and

obtained from the Janssen transgenic rodent facility, Belgium). A separate, satellite group of 6 HO

of 7 months of age were used for immunohistochemical studies.

AppNL-G-F Mice

In vivo data were obtained from 8 HO males AppNL-G-F mice (generated by Saito and colleagues

(Saito et al., 2014a) and obtained from the Janssen transgenic rodent facility, Belgium) and 10 age-


P a g e 30 | 128

matched non-litter mates WT C57BL/6J male mice (Charles River, France). A separate, satellite

group of 36 AppNL-G-F male mice were group-housed in individually ventilated cages for

biochemical and immunohistochemical studies.

3.2. Surgery

Surgeries were carried out when animals were 3 months old. The surgeon was blinded to animal

genotype before carrying out the procedure. Anesthesia was induced by an isoflurane inhalation

(O2, N2O and 5% isoflurane) for 2 minutes and animals inserted into a stereotactic frame

(StereoDrive, Neurostar, Germany). During the surgical procedure, anesthesia was maintained

using a continuous flow of gas (O2, N2O and 2% isoflurane) delivered via an inhalation mask. A

homeothermic blanket system was used to sustain a stable 37-38°C body temperature. To minimize

pain during surgery, a subcutaneous injection of analgesic Piritramide (dipidolor, 0.025mg/kg)

was administered to mice and Metacam (0.1 ml/kg) to rats. As a further precaution to minimize

pain from the surgery, a local spray analgesic (Xylocaine, 10%) was applied at the surgery site.

An incision was made along the sagittal plane to expose the skull, and the scalp held open using

suture thread tied to the stereotactic frame. The skull was cleaned using saline solution and dried

using swabs and cotton buds. Holes were drilled for the placement of recording electrodes; all

coordinates relative to bregma, anterior-posterior (AP), medial-lateral (ML), dorsal-ventral (DV).

Two stainless steel screws were fixed over the left frontal and right occipital lobes to secure the

implant. Depth electrodes consisted of a single fomvar-insulated tungsten wire (100μm diameter

with a blunt-tipped, Peira, Belgium) and surface electrodes were 500μm diameter gold-plated pins,

impedance= 150. Electrodes were placed in different brain structures (Franklin & Paxinos, 1997;

Paxinos & Watson, 1998, for rats and mice, respectively) (see Table 1 for rats and Table 2 for

mice). All electrodes were referenced to the same ground electrode placed on the midline above

the cerebellum (-1.5mm AP, 0.5mm ML) and grounded by a pin positioned on the midline above

the occipital lobe (-5.0mm AP, 0.0mm ML). After placing all electrodes, a multichannel connector

(Nano strip connector, Omnetics, Minneapolis, USA) was affixed using dental cement to the

cranium and the wound sutured around the implant. Animals’ recovery and well-being were

closely monitored for approximately 10 days until they were fully recovered.


P a g e 31 | 128

Brain Structure AP (mm) ML (mm) DV (mm)

frontal associated cortex (Frt Ass Ctx) 5.5 -1.5 0

Left M1/M2 cortex (L-M1/M2 Ctx) 2 -2 0

Right M1/M2 cortex (R-M1/M2 Ctx) 2 2 0

Cingulate cortex (Cg) 1 0 0

Left lateral cortex (Lat Ctx) -3.5 -4 0

Left V1 cortex (V1 Ctx) -7.2 -4 0

Medial prefrontal cortex (mPFC) 3.0 0.7 -3.6

Retrosplenial cortex (RSC) -3.5 0 0

Table 1. Coordinates of surgical implanted electrodes in different brain areas for rats. AP: anterior-

posterior, ML: medial-lateral, DV: dorsal-ventral (according to Paxinos & Watson 1998)

Brain Structure AP (mm) ML (mm) DV (mm)

Dorsal medial striatum (DMS) 1.4 -1.0 -2.5

dorsal CA1 region of the hippocampus

(dCA1)

-1.7 -1.5 -1.7

Cingulate cortex (Cg) -0.5 0.0 0.0

Retrosplenial cortex (RSC) -1.8 0.0 0.0

Table 2. Coordinates of surgical implanted electrodes in different brain areas for mice. AP:

anterior-posterior, ML: medial-lateral, DV: dorsal-ventral (according to Franklin & Paxinos, 1997)

3.3. Behavioral Task

Trial-unique delayed Non-matching-to-location Task

The TUNL task was used for the study conducted with the McGill-R-Thy1-APP rats (see chapter

4). TUNL is a touchscreen task used to assess working memory and spatial pattern separations in

rats (Oomen et al., 2013; Talpos, McTighe, Dias, Saksida, & Bussey, 2010).


P a g e 32 | 128

Apparatus

Rats were trained in operant boxes (modified from Med Associates Inc. Fairfax, Vermont): 2 sides

of the box were constructed of clear acrylic glass. One of the other 2 sides was equipped with a

pellet receptacle containing a light and an infrared nose-poke detector, a tone generator, and a

house light. The remaining side of the box was equipped with a touch-sensitive, flat-screen, LCD

computer monitor (26.5 cm x 40 cm). The monitor was then covered with a “mask”, a piece of

black acrylic glass with 66 small apertures dividing the touchscreen into multiple response fields

in which stimuli were presented. Nose-poking on the screen was detected with an infrared touch

detection system. Boxes were placed in sound attenuated chambers fitted with a small ventilation

fan. The floor of the chamber consisted of aluminum bars spaced approximately 1 cm apart. Each

operant box was controlled by K-limbic software, version 1.20.2 (Conclusive Solutions,

Sawbridgeworth, UK). For the description of the operant box see Figure 4.

Figure 4. Image of the rat touchscreen operant box (modified from Med Associates Inc. Fairfax,

Vermont). One wall of the operant box was equipped with a flat-screen monitor and touchscreen

infrared detection system. The monitor was covered with a black acrylic glass mask. The opposite

side of the box was equipped with a pellet receptacle containing a light and an infrared nose-poke

detector. The other 2 sides of the box were constructed of clear acrylic glass. In the image, a rat is

shown with the head-stage to wirelessly record electrophysiological brain activity during the task.


P a g e 33 | 128

Touchscreen Pre-training Stages

Habituation. Rats were exposed once for 30 minutes to the operant boxes to habituate to the

environment. During this time, the touchscreen was not activated, and food pellets were available

inside the food magazine.

Tone-training. Rats were first trained to make an association between a tone and the delivery of a

food pellet. This association was created by delivering a pellet after the activation of a 0.5 s tone.

To ensure the formation of an association, pellets were not delivered until the previous pellet had

been collected. Once the reward was retrieved, the next trial was automatically initiated. To

advance to the next phase of training, rats had to complete 60 trials within 60 minutes (1

session/day).

Screen-touch training. The next phase of training consisted of building associative strength

between nose-poking the screen and receiving a food pellet. During this phase, nose-poking any

area of the screen was followed by the 0.5 s tone and the delivery of a pellet (all-locations). Rats

were trained in this way until they successfully completed 60 trials in 45 minutes. The second and

last phase of the screen-touch training consisted in a smaller area of the screen being randomly

illuminated where the rats were required to nose-poke to only this location to earn a pellet (one-

location). This process was conducted until rats successfully completed 60 trials in 45 minutes. As

with the tone-training, a new trial would not be started until the previous pellet had been collected.

After this step rats were ready to be advanced to TUNL task training.

TUNL Task Acquisition

Briefly, rats were trained up to 84 trials within a 60 minutes daily session with an inter-trial-interval

(ITI) of 20 seconds on the TUNL task as follows: A session began with the delivery of a food

pellet. Collection of the pellet activated the sample stage where the touchscreen displayed a white

square on the screen, referred as the sample stimulus, where the rat needed to nose-poke. A reward

pellet was delivered in 1 out of 3 trials to maintain robust sample stage initiation. After a variable

delay, the rat activated the choice stage by nose-poking in the food magazine. During this phase,

the rat was presented with two stimuli on the screen: the sample stimulus, and a new one in a novel

location, referred to as the choice stimulus. To obtain a reward, the rat had to select the choice

(novel) stimulus. By manipulating the delay between the sample and choice stage the rat’s ability


P a g e 34 | 128

to remember the sample stimulus to select the novel stimulus during the choice stage was

investigated. During this stage rats were fitted with head-stages used for the LFPs recordings on

alternating test days to acclimatize them to wearing the head-stage. After the acquisition of the

task was achieved (80% correct during variable delays for 2 consecutive days), rats moved into

the delay test.

TUNL Task Delay Test

During this phase, rats were exposed to 2- and 8-second delays in the TUNL task while

electrophysiological recordings were being performed. Initially, rats were trained in the 2 seconds

delay until they completed the task with 80% accuracy for 2 consecutive sessions. Once this was

achieved, rats completed the TUNL task with 2 seconds delay simultaneously with the

electrophysiological recording. Thereafter, the delay was increased to 8 seconds and

electrophysiological measurements were also performed. During the TUNL task, the following

parameters were recorded using the K-limbic software: % correct responses [(correct responses /

(correct + incorrect)) *100]; latency to initiate the sample stage; latency to respond to the sample

stimulus; latency to initiate the choice stage; latency to respond to the choice stimulus, and to

collect the reward. Furthermore, the rats had limited time to perform each stage of the task and if

they exceeded this time the trial ended, and it was considered an omission. For instance, rats had

to respond to the sample stimulus within 30s, to activate the choice stage within 10s, and to respond

to the choice stimulus within 10s, otherwise the trial was finished. Incorrect trials and omissions

were followed by a correction trial (CT). During all CTs the stimulus presentation was in the same

location as the preceding trial, and the trail repeated unit the rats responded to the correct window.

Outcomes during CTs were not included in trial count or any analyzes. For an illustration of the

TUNL task see Figure 5.


P a g e 35 | 128

Figure 5. Trial-Unique Non-match to Location (TUNL) task. A session began with the delivery of

a food pellet in a food magazine opposite to the touchscreen. Collection of the pellet initiated the

trial and activated the sample stage where the sample stimulus appeared on the screen. After nose-

poking on the stimulus, the delay was activated. After a delay (2 seconds in this illustration), the

rat could activate the choice stage by nose-poking in the food magazine. During this phase, the rat

was presented with two stimuli on the screen, the sample stimulus, and a new one in a novel

location, referred to as the choice stimulus. Rats were reinforced with a food pellet when selecting

the choice stimulus and the action was registered as a correct response. If the sample stimulus was

selected, the rat did not receive a reward, the light of the operant chamber went off for 5 seconds,

the response was registered as incorrect, and a CT was initiated. Grey area represents local field

potentials (LFPs) analysis. Epochs of 1.0 seconds immediately preceding activation of the

touchscreen for sample and choice stages during correct responses were selected for

electrophysiological analysis of LFP recordings.

Visual Discrimination Task

The VD task was used for the study conducted with the AppNL-G-F mice (see chapter 5). The VD

task is a relatively simple task where mice learn to consistently respond to one of two visual

stimuli. Learning this type of discrimination is essential for decision making and adaptive

behavior.


P a g e 36 | 128

Apparatus

Similar to the apparatus used for the TUNL task with rats, mice were trained in operant boxes,

however with appropriately smaller dimensions (modified from Med Associates Inc. Fairfax,

Vermont). In this case, the LCD computer monitor was 11.5cm x 19cm. The monitor was then

covered with a mask with two aperture plates dividing the touchscreen into two response fields

(75mm x 75 mm) in which visual stimuli were presented.

Touchscreen Pre-training Stages

The mice were shaped to use touch screens for responding to stimulus images during five pre-

training stages (Horner et al., 2013). All mice began pre-training at 3.5 months of age and

progressed from each stage on an individual basis based on their performance (i.e. reaching a pre-

set performance criterion).

Habituation. The first stage aimed to familiarize the mice with the operant chambers and extractor

fan noise. Mice were placed in the boxes with lights off for 30 minutes and received ten reward

pellets (TestDiet, USA). Mice had to consume all reward pellets during the session in order to

progress to the next stage.

Tone Association. In this stage, mice established an association between a tone and reward

delivery. A reward was delivered with a tone to begin the session. The food magazine light

remained on from reward delivery until collection. The next reward and tone were delivered after

a 30 seconds ITI when the mouse entered the food magazine. Mice were required to complete 60

trials within 60 minutes for 2 consecutive days to advance to the next stage.

Touch Association. Here, mice were encouraged to touch the screen to receive the reward. During

trials, a white square was presented in one of the two response windows for 30 seconds. The

location of the stimulus presentation was pseudo-random between trials. If mice touched the square

during this time a food reward was delivered with a tone and a 10 second ITI initiated. Otherwise,

the stimulus was removed, and the house light switched on for 10 seconds followed by a 10 seconds

ITI. There was no penalty for touching the other response window. Mice were required to complete

a minimum of 35 trials in 45 minutes to advance to the next stage.

Must Touch. During this stage the association between screen touches and rewards was reinforced.

The procedure and success criterion were the same as in the Touch Association stage, however

new trials did not begin automatically after 30 seconds – animals were required to touch the


P a g e 37 | 128

illuminated square before the next trial could begin. Animals were required to complete a

minimum of 35 trials in 45 minutes to advance to the next stage.

Punish Incorrect. The final stage of pre-training introduced a penalty for indiscriminate screen

touches. The trials proceeded as in the Must Touch stage but touches to the non-illuminated

response window triggered a 5 second timeout with the house light on, followed by a 10 seconds

ITI. These trials were recorded as incorrect and followed by a CT. During all CTs the stimulus

presentation was in the same location as the preceding trial, and the trail repeated until the mice

responded to the correct window. Outcomes during CTs were not included in trial count or any

analyzes. Mice progressed from this stage once they could achieve ≥75% correct responses over a

minimum of 30 trials for 2 consecutive days. The number of trials was capped at 80. During this

stage of the pre-training regime mice were fitted with head-stages used for the LFPs recordings on

alternating test days to acclimatize them to wearing the head-stage. After this step mice were ready

to advance to the VD task testing.

VD Task Tests

VD testing began when mice were at an age of 4.5 months. In VD sessions, a pair of images

previously validated (Horner et al., 2013) (Figure 6 a) were presented on screen. Based on a

subject’s counterbalanced group assignment, each image was designated as either the conditioned

stimulus (S+) or the unconditioned stimulus (S-) and the reward-contingency of the image was

kept constant for each mouse across the experiment. Responses to S+ were rewarded with a food

pellet, while responses to S- were recorded as incorrect and resulted in a 5s timeout with the house

light on. As in the Punish Incorrect pre-training stage, incorrect trials were followed by CTs in

which stimulus locations were kept the same. Stimulus presentations occurred in a pseudo-random

location in each trial, never appearing in the same window in more than 3 consecutive trials

(excluding CTs) and total presentations counterbalanced between the two locations. Mice

completed one 45-minute session (max 80 trials) of VD testing daily until they achieved an

acquisition criterion set as 2 consecutive sessions at ≥80% correct responses over a minimum of

30 trials. For an illustration of the VD task see Figure 6, b.


P a g e 38 | 128

Figure 6. Visual discrimination (VD) touchscreen task. a. The spider/plane stimulus image pair

used in the VD (Horner et al., 2013). In this example, the spider was the rewarded image

(conditioned stimulus: S+) and the plane the unrewarded (unconditioned stimulus: S-) image.

Image contingency was counterbalanced across animals. b. Illustration of the VD task. A trial

began with the delivery of a food pellet in the food magazine opposite to the touchscreen.

Collection of the pellet initiated the trial where the two images appeared on the touchscreen. Nose-

poking on the S+ stimulus activated the delivery of a food reward and the trial was registered as a

correct response. Nose-poking the S- stimulus did not delivery a food reward, the light of the

operant chamber went off for 5 seconds, and the response was registered as incorrect. Incorrect

trials were followed by correction trials. Grey area represents LFPs analysis. Epochs of 1.4 seconds

immediately preceding activation of the touchscreen during correct responses were selected for

electrophysiological analysis of LFP recordings.


P a g e 39 | 128

3.4. Electrophysiological Measurements

Recordings

LFPs were sampled with a frequency of 1000 Hz, referenced to the ground electrode placed

midline above the cerebellum, and high-pass filtered above 1 Hz using small-size W4-HS (mice)

and W16-HS (rats) wireless head-stages, connected to a W2100 system (Multichannel systems,

Germany). All analyzes were done with built-in and custom-written routines in Matlab

(MathWorks, 2014a). Detection of artifacts and wrongly placed electrodes was carried out in three

steps. Firstly, data were visually inspected and electrodes with noise were excluded for further

analysis. Secondly, epochs were excluded during the Matlab analysis using an amplitude method

(i.e. artifact rejection) with a SD cut-off of 10. Finally, at the end of the experiment, animals were

deeply anesthetized and electrolytical lesions were produced at the selected recording sites using

a current generator apparatus (500µA for 30 seconds, MC Stimulus II, Multichannel systems,

Germany), then animals were euthanized, and their brain tissue was frozen in dry-ice cooled

methylbutane. Coronal sections of frozen brains were obtained using a cryostat and counterstained

for histological verification of sub-cortical electrode placements.

LFPs were recorded and analyzed in three conditions:

Home-cage Exploration:

Mice at 5 and 8 months and rats at 3.5 months of age were placed for 1 hour in their home-cage

environment with a camera (UI-3140CP-C-HQ Rev.2, IDS Imaging, Germany) above the cage.

The activity recoded with the camera was synchronized with the recorded LFPs. Home-cages were

placed in sound attenuated chambers fitted with a small ventilation fan that also provided a mild

masking background noise. Behavioral activity was scored using Smart software v3.0 (Panlab,

Barcelona) and categorized into different activity levels based on speed. Speed and epoch length

were selected to match behavioral task conditions: only segments where the speed was between 2

and 30 cm/sec for mice and 9 and 50 cm/sec for rats were considered for the analysis. Using a

sliding window approach, we selected LFPs data at 1.4 seconds epochs for mice and 1.0 seconds

for rats for further analysis.


P a g e 40 | 128

TUNL Task:

Neuronal and behavioral data were synchronized with a precision of 10 milliseconds using a

transistor-transistor logic output signal generated by the food magazine light of the operant box.

Recordings of sessions with 2- and 8-seconds delay of the TUNL task were included in the study.

Response latencies were used to select an epoch for analysis within each trial of 1.0 second prior

to correct sample and choice touchscreen responses (Figure 2). This value was used to maximize

the length of the window that contained brain activity related to choose-making immediately prior

to screen-touch while avoiding inclusion of activity related to trial initiation.

VD Task:

Neuronal and behavioral data were synchronized with a precision of 10 milliseconds using a

transistor-transistor logic output signal generated by the food magazine light of the operant box.

Recordings of the first (Task_Start) and last (Task_End) session of the VD task were included in

the study. Response and collection latency data were used to select an epoch for analysis within

each trial of 1.4 seconds prior to a correct touchscreen response (Figure 6, b). This value was used

to maximize the length of the window containing brain activity related to choose-making

immediately prior to screen-touch while avoiding inclusion of activity related to trial initiation.

Relative and Absolute Power Spectral Density Analysis

Welch’s PSD was estimated and analyzed for frequencies ranging from 4 to 200 Hz for each epoch

and condition. For each subject, PSD values were averaged for each examined brain region in 1

Hz frequency bins per session across trials and power was expressed as relative power to the total

power over 4 to 200 Hz. For statistical analysis, relative and absolute PSD data were averaged

across theta (4-12Hz), gamma (30-100 Hz) and HFO (101-200 Hz) (Buzsáki & Draguhn, 2004).

Phase-amplitude Coupling Analysis

The signal was convoluted with complex Morlet wavelets to extract estimates of time-varying

frequency band-specific amplitude and phase from the LFPs data. Theta-gamma PAC was

calculated using an algorithm described previously (Tort et al., 2008). The MI was used to quantify

the modulation of the high frequency amplitude signal (30-200 Hz, estimated in 5Hz steps) by a


P a g e 41 | 128

low frequency phase of the same signal (4-12 Hz, estimated in 0.5 Hz steps). While a MI value

close to zero indicates no relationship between low frequency phase and high frequency amplitude,

a higher value results from stronger phase-to-amplitude modulation. Due to the short duration of

the analysis window (1.0 or 1.4 seconds), within each epoch a lengthiest segment with integer

number of cycles for the considered low frequency phase was extracted, and values of analytical

signals for all the segments within a session were aggregated (over time) in a complex phase space

prior to MI estimations for each animal and electrode. For statistical analysis MI were averaged

across the amplitude frequency low gamma (LG, 30-60 Hz), high gamma (HG, 61-100 Hz) and

HFO (101-200Hz). This gamma sub-banding was done as it has been suggested that different

frequency bands act as different channels for communication. For instance, in the CA1 region of

the hippocampus gamma oscillations split in two different components indicating different origins.

LG arises from interactions with the CA3 region of the hippocampus, where HG is thought to

arises from interactions with the entorhinal cortex (Colgin et al., 2009).

3.5. Brain Aβ1-42 Enzyme-linked Immunoassay (ELISA)

AppNL-G-F mice were euthanized by decapitation after which brains were excised. Olfactory lobes

and hindbrain were removed from tissue processed for biochemistry. Brain hemispheres were

weighed, immediately frozen on dry-ice and stored at -80°C prior to biochemical (right

hemisphere) analysis. For the preparation of brain homogenates, tissue was thawed on ice in pre-

cooled GuHCl extraction buffer (5M Guanidin-hydrochloride, 50 mM Tris-HCl, pH 8.0, 1

mL/100mg tissue). Homogenization was done utilizing 4.5 mL TallPrep tubes (MP Biomedicals)

containing Lysing matrix D (1.4 mm ceramic beads) and FastPrep-24 5G homogenizer (MP

Biomedicals). Subsequently, homogenates were shaken for 3 hours at room temperature and stored

at -80 °C until further use.

Aβ1-42 in AppNL-G-F brain extracts were quantified in a one-step direct sandwich ELISA

with capture antibody JRF/cAβ42/26 that specifically recognizes Aβ ending at amino acid 42, and

detection with SULFOTAG™-labelled 3D6 antibody that recognizes Aβ starting with amino acid

1. Equal volumes of brain homogenates of different mice belonging to the same age group (4 to 6

mice per time point) were pooled and then further diluted 1:10 in casein buffer (0.1% casein in

PBS). Next, samples were spun for 20 minutes at 20.00 g (4 °C) and supernatants were collected


P a g e 42 | 128

to be further diluted in 0.5M GuHCl buffer to optimal dilutions for use in the assay. Human Aβ1–

42 standard peptide (Anaspec, San Jose, CA, USA) was dissolved in DMSO at 0.1 mg/mL and

stored at -80°C. For use in ELISA, a number of standards were diluted in 0.5M GuHCl buffer from

20.00 pg/mL down to 0 pg/mL.

Capture antibody was diluted in PBS (1.5 µg/mL), coated onto Multi-array 96-well

SECTOR plates (Meso Scale Discovery, L15XA, 30 µL/well) and incubated overnight at 4°C.

After five washes with PBS containing 0.5% Tween20 (wash buffer), the plates were blocked in

0.1% casein buffer (2 hours at room temperature while shaking) and washed again with wash

buffer (5x). Prior to their addition to the plates, samples and standards were mixed in an equal

volume of SULFOTAG™-labelled detection antibody (diluted in 0.1% casein buffer) and were

subsequently incubated on the plates overnight at 4°C. The next day, plates were washed five times

with wash buffer, 2 × MSD Read Buffer T was added to the wells and plates were immediately

read with the MESO SECTOR S 600 plate reader. Using Meso Scale software, raw signals were

normalized against the standard curve.

3.6. Immunohistochemistry

Mice and rats were sacrificed by decapitation; brains were removed from the skull, and the left

hemisphere was post-fixed overnight in a formalin-based fixative, embedded in paraffin and sliced

(5 µm) with a microtome. Following deparaffinization and rehydration of the sections, antigen

retrieval was performed (10 minutes incubation in 70% formic acid or heat-induced epitope

retrieval in EDTA buffer) and endogenous peroxidase activity was blocked with 3% hydrogen

peroxide. Samples were incubated overnight at room temperature or 30 minutes at 37°C with

biotinylated primary antibodies (Table 3, Table 4), diluted in antibody diluent with background

reducing components (DAKO, Glostrup, Denmark). After extensive washing, secondary

biotinylated antibody was applied for 30 minutes, followed by streptavidin-HRP (Vector Labs,

Burlingame, CA, USA) and chromogenic labelling with 3,3-diaminobenzidine (DAB, DAKO).

Slides were counterstained with hematoxylin, dehydrated and permanently mounted (Vectamount,

Vector Labs). Imaging was performed with a NanoZoomer slide scanner (Hamamatsu Photonics,

Shizuoka, Japan). For the experiment carried out using the AppNL-G-F, slides were analyzed with

Matlab using Phaedra (Cornelissen, Cik, & Gustin, 2012). Regions-of-interest (ROIs) were


P a g e 43 | 128

manually delineated in accordance with a stereotaxic atlas (Franklin & Paxinos, 1997) and for each

ROI the percentage of DAB-labelled area per total area was calculated.

Primary Ab Company Concentration Secondary Ab Company Concentration

JRD/sAPP/32 In-house 0.3 µg/ml

Monoclonal

rabbit Anti-

mouse/rat IgG1

+ IgG2a + IgG3

Abcam

ab133469 1/500

JRF/cAβ42/26 In-house 1 µg/ml

JRF/AβN/25 In-house 2 µg/ml

McSA1 Abnova 0.5µg/ml

4G8-

biotinylated

Biolegend 1 µg/ml

Table 3. Primary and secondary antibodies for McGill-R-Thy1-APP experiments.

Primary Ab Company Concentration Secondary Ab Company Concentration

JRF/cAβ42/26 In-house 2 µg/ml Anti-mouse/rat

IgG2a Abcam

1/500

JRF/AβN/25 In-house 1 µg/ml 1/250

Table 4. Primary and secondary antibodies for AppNL-G-F experiments.

3.7. McSA1 Antibody Specificity

The commercial antibody McSA1 (Abnova; cat# MAB5669) is commonly used in literature

investigating pathology the McGill-R-Thy1-APP rats (Iulita et al., 2014, 2017; Leon et al., 2010;

Wilson et al., 2016). Although this antibody is reported to specifically bind Aβ with amino acid

epitope 1-12, it is neither clear from publications nor from the vendors website if this antibody is

a neo-epitope antibody (specific for the cleaved Aβ). If it is not, it would cross-react with human

APP or cleaving products containing Aβ1-12 epitope, like sAPPα. Therefore, to investigate the

specificity of McSA1 antibody, a direct coating ELISA was performed with 500ng/ml coating with

recombinant human sAPPα, Aβ1-42, and Aβ 11-42. Control antibodies can be seen in Table 5.

15nM concentration was coated for each protein/peptide.


P a g e 44 | 128

Antibody Epitope Reactivity

sAPPα Aβ1-42 Aβ11-42

JRD/sAPP/32 APP yes no no

JRF/AbN/25 Aβ1-x no yes no

4G8 Aβ 17-24 no yes yes

JRF/Abtot/17 Aβ 1-16 yes yes no

J&JPRD/hAb11/1 Aβ11-x no no yes

JRF/cAb42/26 Aβx-42 no yes yes

Table 5. Control antibodies used to investigate McSA1 antibody specificity.

3.8. Genotyping

Genotyping of McGill-R-Thy1-APP and AppNL-G-F animals was conducted at Charles River (Lyon,

France) using protocols previously described (Leon et al., 2010; Saito et al., 2014a).

Droplet Digital PCR (ddPCR)

To confirm the genotype of the McGill-R-Thy1-APP rats in-house, tails were obtained at the end

of the experiments and genomic DNA was extracted using DNeasy Blood & Tissue kit (Catalog

Number: QIAG69506, QIAGET) according to the manufacturer’s instructions. The QX200

ddPCR system (Bio-Rad) was used to provide absolute measurements on genomic DNA copy

number with high accuracy, precision, resolution and sensitivity.

3.9. Statistical Analysis

Statistical analyzes for in vivo data were conducted using JMP®, version 12 (SAS Institute Inc,

NC, US). For statistical inference on behavioral data, the distributions of dependent variables

were assessed for meeting the assumption of normality required in parametric statistical tests.

Group comparisons were made using a two-tailed two-sample t-test. If the data did not meet the

normality assumptions, Wilcoxon rank-sum test (statistics are reported as Z) was used. A mixed-

effect model for repeated measures with genotype, and session as fixed effects and subject as a


P a g e 45 | 128

random effect was used to investigate percentage of correct responses during the VD task in the

AppNL-G-F experiment. Residuals of models were inspected for normality and an alpha level of 0.05

was used to determine significance in all statistical tests. For statistical inference on

electrophysiological data, most data were non-normally distributed and nonparametric tests were

applied. For within-subject analysis the delta value between the two conditions (start and end of

the VD task) was calculated and compared to zero. Significance was tested using the Wilcoxon

rank-sum test. Correction for multiple comparison was performed using the Benjamini-Hochberg

procedure with a false discovery rate (FDR) threshold of q = 0.05 (Benjamini & Hochberg, 1995;

Glickman, Rao, & Schultz, 2014). For clear representation, descriptive statistics, such as mean,

median, standard deviation (SD), and ranges, together with p values and thresholds are presented

in tables and box plots with individual points for each subject, rather than commonly reported bars

graphs (Weissgerber, Milic, Winham, & Garovic, 2015). Box plots represent the interquartile

range; the solid line inside the box indicates the median, the box represents 50% of data points

between the first and third quartile, and the upper and lower whiskers represent scores outside the

middle 50%.

Statistical analyzes for biochemical and immunohistochemical data were conducted using

GraphPad Prism, version 8 (GraphPad Software, Inc.). Using the Shapiro-Wilk test, the

distributions of dependent variables were assessed for meeting the assumption of normality

required in parametric statistical tests. Multiple group comparisons were made using the Brown-

Forsythe and Welch’s Analysis of variance (ANOVA) with Dunnett’s T3 multiple comparisons

test. Data were considered significant when p < 0.05 (1 symbol p < 0.05, 2 symbols, p < 0.01, 3

symbols p < 0.001, 4 symbols p < 0.0001).

P a g e 46 | 128

McGill-R-Thy1-APP Rats Characterization

P a g e 47 | 128

Behavioral, Electrophysiological and Histopathological Characterization of

McGill-R-Thy1-APP Rats

4.1. Introduction

The primary goal of this part of the thesis was to investigate electrophysiological readouts in

combination with a cognitive task to illuminate functional differences between McGill-R-Thy1-

APP (HET and HO) and WT littermate controls. The McGill-R-Thy1-APP model is a transgenic

rat model overexpressing the human APP carrying both the Swedish double mutation (K670N,

M671L) and the Indiana mutation (V717F). Pathologically, these rats were reported to show

already at 1 week of age intraneuronal Aβ accumulation in different cortical and hippocampal

areas. Extracellular Aβ deposits in the subiculum and entorhinal cortex start to become appearing

after 6 months in HO rats. By 13 months, plaque pathology spreads across the hippocampus and

cortical regions (Leon et al., 2010). At the level of cognition, the McGill-R-Thy1-APP rat starts to

show cognitive impairment in spatial memory and VD at around 3 months of age (Galeano et al.,

2014; Leon et al., 2010; Wilson et al., 2016). Furthermore, at the electrophysiological level, this

transgenic rat shows resistance to long term potentiation (LTP) induction in vivo and in vitro in

the dorsal hippocampus during the pre-plaque stage at approximately 3.5 months of age (Qi et al.,

2014). At the network level, the McGill- R-Thy1-APP rat was described to show a progressive

functional decline (Parent et al., 2017). As with the previous mentioned measurements, the

abnormalities at the connectivity level appear before Aβ plaques, suggesting that human Aβ

oligomeric peptides and aggregates exert toxic effects before the formation of plaques.

Rats had to perform in the TUNL task with a 2- and 8- seconds delay at 3.5 months of age

while LFPs in different cortical and sub-cortical brain areas were recorded using a wireless

neurophysiological signal acquisition system. In parallel with these recordings, we measured LFPs

at the same age during home-cage exploration, without the behavioral task to investigate neuronal

network changes exclusively associated with pathology. The electrophysiological analysis focused

on PAC between theta (4 – 12 Hz) and gamma (30 – 100 Hz) activity during three distinct

recording conditions related to different cognitive load: the sample and the choice stages of the

TUNL task, and exploration of the home-cage environment. The age of 3.5 months of was selected

because it matches the preclinical AD phase according to the available data. Immunohistochemical


P a g e 48 | 128

analysis was conducted to correlate Aβ plaques deposition with the electrophysiological and

behavioral results.

4.2. Results

Performance during TUNL Task

Pre-training:

During the pre-training stage there were no genotype differences in the number of sessions required

to learn the tone association (WT: M = 1, SD = 0, n = 16; HET: M =1, SD = 0, n = 6; HO: M = 1,

SD = 0, n = 6, Figure 7, left panel), the all-locations (WT: M =2.38, SD = 0.50, n = 16; HET: M =

2.00, SD = 0.00, n = 6; HO: M = 2.33, SD = 0.52, n = 6, Figure 7, middle panel), and the one-

location (WT: M = 1.13, SD = 0.34, n = 16; HET: M = 1.00, SD = 0.00, n = 6; HO: M = 1.67, SD

= 0.82, n = 6, Figure 7, right panel) stages of touchscreen pre-training.

Figure 7. Pre-training of the Trial-unique Non-match to Location (TUNL) task in WT, HET, and

HO McGill-R-Thy1-APP rats. Number of days to acquire the tone-training (left panel), all-

locations (middle panel), and one-location (right panel) of the screen-touch training are shown.

Performance during the different pre-training stages did not differ between genotypes. Data are

represented in box plots with individual points for each subject. Statistically significance level at

p < 0.05. ns: non-significant.


P a g e 49 | 128

TUNL Task Acquisition:

McGill-R-Thy1-APP HO failed to acquire the TUNL task as they had a high level of omissions

(HO: M = 96.51%, SD = 5.45, n = 6, Figure 8). We compared body weight between genotypes to

assess if the lack of performance on the HO rats was associated with food reward motivation, but

no differences were observed (Figure 9). Furthermore, HO performed similarly to the WT and

HET during the pre-training phase of the task, demonstrating motivation to respond in the operant

box. Due to the low number of trials, HO rats were not included in further behavioral and

electrophysiological analyzes during the task.

Figure 8. Percentage of omissions during acquisition of the Trial-unique Non-match to Location

(TUNL) task in WT, HET, and HO McGill-R-Thy1-APP rats. HO rats had a significantly higher

number of omissions that WT and HET rats. Data are represented in box plots with individual

points for each subject. Statistically significance level at p < 0.05. ns: non-significant.

Figure 9. Mean body weight (g) during experiments in WT, HET, and HO McGill-R-Thy1-APP

rats. No differences were observed between genotypes. Data are represented in box plots with

individual points for each subject. Statistically significance level at p < 0.05. ns: non-significant.


P a g e 50 | 128

TUNL Task Delay Test

From the original sample size, 7 WT and 1 HET rats were excluded from the analysis due to

insufficient electrophysiological signal quality or problems with the implant. An extra WT rat had

surgical complications before the 8 seconds delay and was removed from the further studies.

Finally, 2 WT and 1 HET did not reach behavioral performance criterion to be included in the

delay test. For the rats included on the analysis, there were no differences in performance during

the 2- and 8- seconds delay between WT and HET rats.

2 seconds delay. Both genotypes performed well above change levels and they did not

differ during sessions with LFPs recordings (WT: M = 86.30, SD = 5.45, n = 7; HET: M = 83.49,

SD = 6.94, n = 4; Z = -0.66, p = 0.51 Figure 10, a). Analysis of percentage of CTs (WT: M =14.20,

SD = 8.34, n = 7; HET: M = 17.04, SD = 6.34, n = 4; Z = 0.47, p = 0.64 Figure 10, b), and percentage

of omissions (WT: M = 10.64, SD = 11.74, n = 7; HET: M = 5.21, SD = 5.69, n = 4; Z = -0.094, p

= 0.925 Figure 10, c) did not identify a significant difference between genotypes. Furthermore, no

differences were found between WT and HET rats for sample response (WT: M = 4.50, SD = 2.47,

n = 7; HET: M = 2.93, SD = 0.38, n = 4; Z = -0.28, p = 0.78 Figure 10, d, first panel), choice

initiation (WT: M = 2.09, SD = 0.86, n = 7; HET: M = 3.01, SD = 0.40, n = 4; Z = 1.42, p = 0.16

Figure 10, d, seconds panel), choice response (WT: M = 3.29, SD = 0.63, n = 7; HET: M = 2.70,

SD = 0.34, n = 4; Z = -1.60, p = 0.11 Figure 10, d, third panel), and reward collection (WT: M =

1.47, SD = 0.57, n = 7; HET: M = 1.34, SD = 0.18, n =; Z = 0.09, p = 0.92 Figure 10, d, fourth

panel) latencies.

8 seconds delay. As with the 2 seconds delay, no differences were found between WT and

HET rats. Analysis of percentage of correct responses (WT: M = 77.99, SD = 10.63, n = 6; HET:

M = 70.33, SD = 3.07, n = 4; Z = -1.18, p = 0.24, Figure 11, a), CTs (WT: M = 22.62, SD = 9.55,

n = 6; HET: M = 28.57, SD = 4.86, n = 4; Z = 0.97, p = 0.33, Figure 11, b), and omissions (WT:

M = 4.52, SD = 5.25, n = 6; HET: M = 1.23, SD = 0.82, n = 4; Z = -1.07, p = 0.28, Figure 11, c)

did not identify a significant difference between the two groups. Latency to sample response (WT:

M = 4.30, SD = 0.53, n = 6; HET: M = 4.14, SD = 1.27, n = 4; Z = 0.32, p = 0.75, Figure 11, d,

first panel), choice initiation (WT: M = 1.13, SD = 0.23, n = 6; HET: M = 1.22, SD = 0.26, n = 4;

Z = 0.32, p = 0.75, Figure 11, d, second panel), choice response (WT: M = 3.83, SD = 0.51, n = 6;

HET: M = 3.68, SD = 0.68, n = 4; Z = -0.11, p = 0.91, Figure 11, d, third panel), and reward


P a g e 51 | 128

collection (WT: M = 1.12, SD = 0.12, n = 6; HET: M = 1.17, SD = 0.12, n = 4; Z = 0.96, p = 0.34,

Figure 11, d, fourth panel) did not differ between genotypes.

Figure 10. Results of the Trial-unique Non-match to Location (TUNL) task during the 2 seconds

delay in WT and McGill-R-Thy1-APP, HET rats. Percentage of correct responses (a), percentage

of correction trials (b) and percentage of omissions (c) did not differ between genotypes. d.

latencies in seconds (sec) to respond to the sample stimulus (first panel), choice initiation (second

panel), choice response (third panel), and reward collection (fourth panel) were not different

between WT and HET rats. Data are represented in box plots with individual points for each

subject. Statistically significance level at p < 0.05, ns: non-significant


P a g e 52 | 128

Figure 11. Results of the Trial-unique Non-match to Location (TUNL) task during the 8 seconds

delay in WT, McGill-R-Thy1-APP and HET rats. Percentage of correct responses (a), percentage

of correction trials (b) and percentage of omissions (c) did not differ between genotypes. d.

latencies in seconds (sec) to respond to the sample stimulus (first panel), choice initiation (seconds

panel), choice response (third panel), and reward collection (fourth panel) were not different

between WT and HET rats. Data are represented in box plots with individual points for each

subject. Statistically significance level at p < 0.05, ns: non-significant

Brain Oscillations Analysis During TUNL Task

Power Spectral Density

Before investigating the PSD in the 8 brain areas of interest during sample and choice stages of

the TUNL task, we grouped frequencies into two bands: theta (4 to 12 Hz) and gamma (30 to 100

Hz), based on previously proposed guidelines (Jobert et al., 2012). Analysis of PSD between WT

and HET rats revealed a significant increase for HET in the mPFC for both sample and choice

stages at the 2- and 8- seconds delay in the gamma band (Figure 12, Figure 13, c and d, and Table

6). Furthermore, theta band for HET was higher than for WT during the sample stage of the 2

seconds delay (Figure 12, c). A within-subject analysis between sample and choice stages revealed

no significant difference in PSD for any of the frequency bands for both genotypes at the 2 seconds

delay (Table 7) for complete descriptive statistics. Results for the 8 seconds delay were similar as

those for the 2 seconds delay (data not presented).


P a g e 53 | 128

Figure 12. Mean power spectral density (PSD) across animals during the Trial-Unique Non-match

to Location (TUNL) task for the medial prefrontal cortex (mPFC) during the 2 seconds delay test.

Mean ± 95% confidence interval PSD in the 4 to 100 Hz band for WT (black curve) and HET

(green curve) rats for the (a) sample and (b) choice stages of the TUNL task. HET rats had a

significantly higher PSD at the gamma band (right panels) than WT rats for sample (c) and (d)

choice stages. Additionally, PSD at the theta band for HET rats was significantly higher than WT

for the sample stage. Data are represented in box plots with individual points for each subject.

Significance level at q = 0.05, false discovery rate (FDR) corrected for multiple comparisons. ns:

non-significant.


P a g e 54 | 128

Figure 13. Mean power spectral density (PSD) across animals during the Trial-Unique Non-match

to Location (TUNL) task for the medial prefrontal cortex (mPFC) during the 8 seconds delay test.

Mean ± 95% confidence interval PSD in the 4 to 100 Hz band for WT (black curve) and HET

(green curve) rats for the (a) sample and (b) choice stages of the TUNL task. HET rats had a

significantly higher PSD at the gamma band (right panels) than WT rats for sample (c) and (d)

choice stages. Data are represented in box plots with individual points for each subject.


non-significant.


P a g e 55 | 128

Table 6. Between-subject analysis of power spectral density (PSD) during the Trial-unique Non-

match to Location (TUNL) during the 2 seconds delay test.

Conditions: Sample Stage, Choice Stage.

Brain areas: Cingulate Cortex (Cg), frontal associated cortex (Frt Ass Ctx), V1 cortex (V1 Ctx),

lateral cortex (Lat Ctx), left and right M1/M2 cortex (L-M1/M2 Ctx, R-M1/M2 Ctx), retrosplenial

cortex (RSC), medial prefrontal cortex (mPFC).

Frequencies: theta and gamma.

Z: Wilcoxon rank-sum test.

Significance level at q = 0.05, false discovery rate (FDR) corrected for multiple comparisons.

Condition Brain Area Freq Z Prob > [Z] p value rank FDR threshold

Theta 0,472 0,637 1,000 0,008

Gamma 0,283 0,777 3,000 0,025

Theta 0,000 1,000 3,000 0,025

Gamma 0,094 0,925 2,000 0,017

Theta -1,417 0,156 2,000 0,017

Gamma -0,661 0,508 4,000 0,033

Theta -0,850 0,395 1,000 0,008

Gamma -0,472 0,637 2,000 0,017

Theta 0,283 0,777 2,000 0,017

Gamma 0,472 0,637 1,000 0,008

Theta -0,283 0,777 2,000 0,017

Gamma 0,000 1,000 4,000 0,033

Theta 0,094 0,925 4,000 0,033

Gamma 1,417 0,156 2,000 0,017

Theta 2,362 0,018 3,000 0,025

Gamma 2,551 0,011 2,000 0,017

Theta 0,094 0,925 4,000 0,033

Gamma 0,283 0,777 2,000 0,017

Theta 0,094 0,825 1,000 0,008

Gamma 0,000 1,000 4,000 0,033

Theta -1,417 0,156 1,000 0,013

Gamma -1,039 0,299 3,000 0,025

Theta -0,283 0,777 3,000 0,025

Gamma -0,094 0,925 4,000 0,033

Theta 0,000 1,000 4,000 0,033

Gamma 0,283 0,777 3,000 0,025

Theta -0,283 0,777 1,000 0,008

Gamma -0,094 0,925 3,000 0,025

Theta 0,472 0,637 3,000 0,025

Gamma 1,417 0,156 1,000 0,008

Theta 1,984 0,047 4,000 0,033

Gamma 2,551 0,011 1,000 0,013

R-M1/M2 Ctx

RSC

mPFC

R-M1/M2 Ctx

RSC

mPFC

Sample Stage

Choice Stage

Cg

L-Frt Ass Ctx

V1 Ctx

Lat Ctx

L-M1/M2 Ctx

Cg

L-Frt Ass Ctx

V1 Ctx

Lat Ctx

L-M1/M2 Ctx


P a g e 56 | 128

Genotype Brain Area Freq Condition Mean SD Median Rangle

Theta 69.43 53.12 47.93 210.77

Gamma 5.47 3.15 4.70 29.71

Theta 75.53 62.22 49.88 221.77

Gamma 5.57 3.82 4.58 48.84

Theta 92.53 65.74 86.04 309.31

Gamma 5.88 3.29 5.29 17.06

Theta 88.43 72.10 65.45 294.53

Gamma 5.89 3.45 5.10 20.68

Theta 65.72 56.53 52.36 321.18

Gamma 4.70 2.6 4.22 35.98

Theta 84.72 143.94 47.63 985.37

Gamma 4.89 3.22 4.13 43.50

Theta 68.11 41.96 58.37 204.04

Gamma 4.96 2.49 4.51 17.80

Theta 66.50 39.78 56.19 173.91

Gamma 5.24 3.44 4.25 25.86

Theta 48.87 39.63 43.67 213.76

Gamma 3.41 2.67 2.45 18.98

Theta 46.62 36.73 40.793 200.04

Gamma 34.58 3.36 2.59 33.79

Theta 30.60 26.16 25.53 135.17

Gamma 2.66 1.67 2.15 10.22

Theta 27.46 21.84 24.92 100.53

Gamma 2.72 1.99 2.08 10.51

Theta 71.83 66.65 49.48 267.26

Gamma 5.26 3.62 4.29 24.21

Theta 67.05 63.79 46.99 261.78

Gamma 5.13 4.13 3.87 44.09

Theta 66.89 57.15 49.83 285.41

Gamma 5.15 3.60 3.84 19.15

Theta 65.15 58.11 44.70 258.82

Gamma 4.97 3.89 3.77 21.79

Theta 50.43 33.05 42.74 131.68

Gamma 5.83 2.98 5.22 25

Theta 52.90 37.31 42.06 131.53

Gamma 6.00 3.44 5.33 36.10

Theta 57.84 32.64 51.62 162.67

Gamma 6.53 2.89 6.00 15.06

Theta 57.53 34.70 47.43 151.06

Gamma 6.61 3.02 6.22 19.25

Theta 58.07 40.16 48.19 190.57

Gamma 6.27 3.47 5.57 24.38

Theta 61.85 47.55 47.84 186.76

Gamma 6.45 4.09 5.52 41.12

Theta 67.70 52.77 54.14 236.57

Gamma 6.06 2.069 5.63 13.39

Theta 66.98 57.78 40.80 231.39

Gamma 6.13 2.88 5.79 16.71

Theta 85.65 86.89 42.68 341.35

Gamma 5.08 3.66 3.95 27.79

Theta 90.65 90.71 42.39 334.64

Gamma 5.23 4.26 3.81 26.62

Theta 99.86 93.02 53.69 345.49

Gamma 6.73 3.55 5.75 17.19

Theta 99.05 106.23 48.00 370.35

Gamma 6.96 4.18 5.62 24.24

Theta 90.83 58.05 73.08 254.70

Gamma 7.26 5.31 5.75 33.11

Theta 93.71 64.62 70.02 252.55

Gamma 7.52 5.78 5.85 41.40

Theta 192.93 103.14 181.48 399.23

Gamma 18.27 10.26 18.25 62.91

Theta 186.92 113.93 159.89 555.89

Gamma 17.58 10.38 16.87 76.34

WT

McGill-R-Thy1-APP

(HET)

WT

McGill-R-Thy1-APP

(HET)

WT

McGill-R-Thy1-APP

(HET)

WT

McGill-R-Thy1-APP

(HET)

mPFC

WT

McGill-R-Thy1-APP

(HET)

WT

McGill-R-Thy1-APP

(HET)

WT

McGill-R-Thy1-APP

(HET)

WT

McGill-R-Thy1-APP

(HET)

L-M1/M2 Ctx

L-Frt Ass Ctx

V1 Ctx

Lat Ctx

R-M1/M2 Ctx

RSC

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Sample Stage

Choice Stage

Cg

Sample Stage

Choice Stage

Sample Stage

Choice Stage


P a g e 57 | 128

Table 7. Power spectral density (PSD, µV2/Hz) during the Trial-unique Non-match to Location

(TUNL) during the 2 seconds delay test.







Conditions: Sample and Choice Stages

Phase-amplitude Coupling

Similarly to PSD analysis, we investigated whether the modulation index (MI) was different

between WT and HET rats during the TUNL task for the 2- and 8- seconds delay. We also

investigated if the modulation changed between sample and choice stages. For this analysis, we

grouped amplitude frequencies in two bands: LG (30 – 60 Hz) and HG (61 – 100 Hz). This banding

was selected as it has been suggested that different frequency bands act as different channels for

communication. A visual inspection of the different phase-amplitude comodulograms during the

2 seconds delay indicated that PAC was present in the RSC (Figure 14) for both genotypes and

stages of the task. Other brain areas did not demonstrate a clear coupling between theta and gamma

for the investigated condition (see Cg as an example, Figure 15). When investigating the difference

in the MI between genotypes, no differences were observed for any of the brain areas (Figure 14,

b Figure 15, b, Table 8). It should be noted that some of the mean phase-amplitude comodulograms

might visually suggest changes between genotypes, but as can be seen in the box plots, some of

these measurements had a high level of variability. Next, we investigated MI differences between

the sample and choice stages within each genotype. For this, we compared the MI for the two

amplitude bands for WT and HET rats and corrected for multiple comparisons using FDR. Table

9 exemplifies that no significant differences were observed between sample and choice for any of

the brain areas and frequency bands. See Table 9 for completed descriptive statistics.

Results for the 8 seconds delay were similar as those for the 2 seconds delay, i.e. no

differences between genotypes or task stages were observed (data not presented).


P a g e 58 | 128

Figure 14. Mean phase-amplitude coupling (PAC) across animals during the Trial-Unique Non-

match to Location (TUNL) for the retrosplenial cortex (RSC) at the 2 seconds delay. a. Phase-

amplitude comodulograms plotted for the sample stage (top panels) and the choice stage (lower

panels) for WT (left) and HET (right) rats. b. Modulation index (MI) between WT and HET rats

for the sample stage (top) and the choice stage (bottom) for low gamma (LG, left) and high gamma

(HG, right). A comparison at each frequency band showed no significant differences. Values

represented in box plots are individual points that indicate the absolute values for each subject.


non-significant.

Figure 15. Mean phase-amplitude coupling (PAC) across animals during the Trial-Unique Non-

match to Location (TUNL) for the cingulate cortex (Cg) at the 2 seconds delay. a. Phase-amplitude

comodulograms plotted for the sample stage (top panels) and the choice stage (lower panels) for

WT (left) and HET (right) rats. B. Modulation index (MI) between WT and HET rats for the

sample stage (top) and the choice stage (bottom) for low gamma (LG, left) and high gamma (HG,


P a g e 59 | 128

right). A comparison at each frequency band showed no significant differences. Values are

represented in box plots were individual points indicate absolute values for each subject.


non-significant.

Table 8. Between-subject analysis of phase-amplitude coupling (PAC) during the Trial-unique

Non-match to Location (TUNL) during the 2 seconds delay test.

Conditions: Sample Stage, Choice Stage.







Condition Brain Area Freq Z Prob > [Z] p value rank FDR threshold

LG 0,4724 0,6366 2 0,025

HG 0,85042 0,3951 4 0,05

LG 1,2283 0,2193 1 0,0125

HG 0,2847 0,7768 3 0,0375

LG -1,0394 0,2986 1 0,0125

HG 0,09449 0,9247 2 0,025

LG 0,28347 0,7768 3 0,0375

HG 0,47246 0,6366 2 0,025

LG 1,41737 0,1564 1 0,0125

HG -0,66144 0,5083 4 0,05

LG 0,28347 0,7768 3 0,0375

HG -1,2283 0,2193 1 0,0125

LG 1,41737 0,1564 1 0,0125

HG 0,85042 0,3951 3 0,0375

LG -0,09449 0,9247 3 0,0375

HG -1,98431 0,0472 1 0,0125

LG -1,0394 0,298 1 0,0125

HG 0,47246 0,6366 3 0,0375

LG 0,28347 0,7768 4 0,05

HG 0,47246 0,6366 2 0,025

LG 0 1 4 0,05

HG 0,094 0,9247 3 0,0375

LG 0 1 4 0,05

HG 1,0394 0,29886 1 0,0125

LG -0,85 0,3951 3 0,0375

HG 1,0394 0,2568 2 0,025

LG -0,85042 0,3951 2 0,025

HG 0 1 4 0,05

LG 1,417 0,1564 2 0,025

HG 0,66144 0,5083 4 0,05

LG -1,7953 0,0588 2 0,025

HG -0,09449 0,9247 4 0,05

RSC

mPFC

Cg

L-Frt Ass Ctx

V1 Ctx

Lat Ctx

L-M1/M2 Ctx

R-M1/M2 Cx

Sample Stage

Choice Stage

Cg

L-Frt Ass Ctx

V1 Ctx

Lat Ctx

L-M1/M2 Ctx

R-M1/M2 Ctx

RSC

mPFC


P a g e 60 | 128

Table 9. Phase-amplitude coupling (PAC) during the Trial-unique Non-match to Location (TUNL)

during the 2 seconds delay test.








P a g e 61 | 128

Brain Oscillation Analysis During Home-Cage Environment Exploration

Power Spectral Density

Our next analysis was conducted to determine if we could observe differences between genotypes

in brain oscillatory activity when rats were exploring the home-cage environment at an age of 3.5

months. In this analysis not only WT and HET rats were included, but also HO rats, as the analysis

was independent of task performance. Contrary to PSD results obtained during the TUNL task, we

did not observed differences between genotypes for any of the frequency bands (Figure 16, Table

10) for the mPFC. Furthermore, as with the PSD during the TUNL task, we did not observe

differences for any of the other brain regions of interest (Table 10).

Figure 16. Mean power spectral density (PSD) across animals during the home-cage exploration

for the medial prefrontal cortex (mPFC). a. Mean ± 95% confidence interval PSD in the 4 to 100

Hz band for WT (black curve), HET (green curve), HO (blue curve) rats at 3.5 months of age

during the home-cage exploration. b. Box plots for theta (left panel) and gamma (right panel),

where individual points indicate PSD for each subject. Significance level at q = 0.05, false

discovery rate (FDR) corrected for multiple comparisons. ns: non-significant.


P a g e 62 | 128

Table 10. Between-subject analysis of power spectral density (PSD) during the home-cage

environment exploration.

Conditions: 3.5 months of age.







Phase-Amplitude Coupling

As with the PSD, no significant differences were observed between genotypes during the home-

cage environment exploration (Table 11) for any of the brain areas at 3.5 months of age.


P a g e 63 | 128

Table 11. Between-subject analysis of phase-amplitude coupling (PAC) during the home-cage

environment exploration.

Conditions: 3.5 months of age.







Pathology

Immunohistochemistry

Aβ accumulation was studied by immunohistochemical evaluation of McGill-R-Thy1-APP

forebrain tissue with multiple antibodies. In-house antibodies (JRF/AβN/25, Figure 17 and

JRF/cAβ42/26, Figure 18) indicated that HO McGill-R-Thy1-APP rats had no Aβ pathology at 7

months of age (panels e and f), while the 2 positive controls (panels b and c) had dense plaque


P a g e 64 | 128

depositions. These results were confirmed with commercial antibody, 4G8 (Figure 19). As these

results were in contradiction with previous reports(Iulita et al., 2014, 2017; Leon et al., 2010;

Wilson et al., 2016), we evaluated the same forebrain tissue samples with the antibody used in

these publications (McSA1, Figure 20). Immunohistochemical results with the McSA1 suggested

limited intracellular Aβ in pyramidal neurons (Figure 17, e and f). Finally, we assessed the samples

with an in-house antibody specific for human and rodent App, JRD/sAPP/32 (Figure 18). As

expected, negative and positive controls, as well as HO McGill-R-Thy1-APP showed positive

staining for App. Altogether, our data indicate that at 7 months of age the HO McGill-R-Thy1-

APP rats do not have Aβ accumulation. Furthermore, our data questions the previously reported

(Grant, Ducatenzeiler, Szyf, & Cuello, 2000) specificity of McSA1 for human Aβ.

Figure 17. Immunohistochemical analysis of Aβ accumulation using an N-terminal antibody

(JRF/AβN/25). a. forebrain tissue of a WT (C57BL/6) mouse at 14 months of age as a negative

control. b and c. forebrain tissue of two transgenic mouse models overexpressing human APP at

14 and 15 months of age, respectively, as positive controls. d. WT McGill-R-Thy-APP at 7 months

of age as a negative control. e and f. two HO McGill-R-Thy-APP at 7 months of age. Scale bars

in the photographs are 2.5 mm and 50 µm (hippocampus inset).


P a g e 65 | 128

Figure 18. Immunohistochemical analysis of Aβ accumulation using a C-terminal antibody

(JRF/cAβ42/26). a. forebrain tissue of a WT (C57BL/6) mouse at 14 months of age as a negative


14 and 15 months of age respectively as positive controls. d. WT McGill-R-Thy-APP at 7 months



Figure 19. Immunohistochemical analysis of Aβ accumulation using the anti-Aβ, 17-24, 4G8

antibody. a. forebrain tissue of a WT (C57BL/6) mouse at 14 months of age as a negative control.

b and c. forebrain tissue of two transgenic mouse models overexpressing human APP at 14 and 15

months of age respectively as positive controls. d. WT McGill-R-Thy-APP at 7 months of age as

a negative control. e and f. two HO McGill-R-Thy-APP at 7 months of age. Scale bars in the

photographs are 2.5 mm and 50 µm (hippocampus inset).


P a g e 66 | 128

Figure 20. Immunohistochemical analysis of Aβ accumulation using the McSA1 antibody. a.

forebrain tissue of a WT (C57BL/6) mouse at 14 months of age as a negative control. b and c.

forebrain tissue of two transgenic mouse models overexpressing human APP at 14 and 15 months

of age respectively as positive controls. d. WT McGill-R-Thy-APP at 7 months of age as a negative

control. e and f. two HO McGill-R-Thy-APP at 7 months of age. Scale bars in the photographs are

2.5 mm and 50 µm (hippocampus inset).

Figure 21. Immunohistochemical analysis of Aβ accumulation using an anti-APP antibody

(JRD/sAPP/32). a. forebrain tissue of a WT (C57BL/6) mouse at 14 months of age as a negative


14 and 15 months of age respectively as positive controls. d. WT McGill-R-Thy-APP at 7 months




P a g e 67 | 128

McSA1 Antibody Specificity Analysis

To further evaluate the specificity of McSA1 for human Aβ, a direct coating ELISA was performed

with a 15 nM concentration coating with recombinant human sAPPα, Aβ1-42, and Aβ 11-42.

Although all control antibodies reacted with the expected coated protein/peptide (Figure 19

b,c,d,e,f, and g), McSA1 antibody reacted not only with Aβ1-42 but also with sAPPα, implying

this antibody is not a neo-epitope antibody (Figure 19,a). Furthermore, the lack of reactivity with

Aβ11-42 indicates its epitope locates in the first 11 amino acids.

Figure 22. Direct coating ELISA with 15nM concentration coating with recombinant human

sAPPα, Aβ1-42, and Aβ 11-42. a. Evaluation of the specificity of McSA1 for human Aβ. McSA1

antibody reacted not only with Aβ1-42, but also with sAPPα, implying this antibody is not a neo-

epitope antibody (specific for the cleaved Aβ). b. Control antibody JRD/sAPP/32 binds to epitope

in sAPPα but not Aβ, therefore it only reacted with sAPPα. c. Control antibody JRF/AbN/25 is a


P a g e 68 | 128

neo-epitope antibody for Aβ 1-x and thus should react with Aβ 1-42, but not with sAPPα nor with

Aβ 11-42. Results indicated binding to Aβ 1-42, but also partial binding to sAPPα. d. Control

antibody JRF/Abtot/17 binds with Aβ epitope 1-12, therefore reacted with sAPPα, and Aβ 1-42.

e. Control antibody 4G8 binds with epitope 17-24, thus reacted with Aβ 1-42 and Aβ 11-42. f.

J&JPRD/hAb11/1 negative control. g. Control antibody JRF/cAb42/26 is a neo-epitope antibody

for Aβ x-42, thus reacted with both Aβ 1-42 and Aβ 11-42, but not with sAPPα.

Genotyping – ddPCR

It is important to notice that the original study design was planned to only include WT and HO

rats, as it has been reported that HET McGill-R-Thy1-APP rats have a low level of pathology

(Leon et al., 2010). After observing that approximately half of the HO rats were not performing in

the TUNL task, while the other half was learning the task, we re-genotyped the rats at the end of

the study using a technology that produces accurate and precise digital PCR: the ddPCR.

Unexpectedly, results indicated that 50% of the HO rats, were actually HET. The re-assigned

genotypes matched clearly with the bimodal performance distribution, as HO rats failed to acquire

the TUNL task, while HET rats performed similarly to WT rats.

4.3. Conclusion

The purpose of this part of the thesis was to characterize the McGill-R-Thy1-APP rats by

performing a behavioral, electrophysiological, and pathological investigation at 3.5 months of age.

In has been suggested that Aβ amyloidosis causes altered neuronal oscillatory activity (Busche &

Konnerth, 2015; Palop & Mucke, 2016). In this chapter, we tested the hypothesis that Aβ

amyloidosis causes aberrant network activity at the preclinical stage of the disease and underlies

the early cognitive disturbances observed in models of AD pathology using the McGill-R-Thy1-

APP rat.

The McGill-R-Thy1-APP HO rats showed a strong impairment in the acquisition of the

TUNL task, as they had a high level of omissions. We excluded demotivation as the main reason

for this impairment, as HO rats performed equally well as the WT and HET during the pre-training

stage of the task. Furthermore, no bodyweight differences were observed between genotypes.

Importantly, a previous study of the HO McGill-R-Thy1-APP rats in the operant touchscreen

apparatus, using a different task than the one here, reported that these rats took over 10 seconds to


P a g e 69 | 128

make touchscreen responses (Wilson et al., 2016). In our experimental designed rats had 30

seconds to respond to sample stimulus, and 10 seconds to activate the choice stage or respond to

the choice stimulus. Therefore, a plausible explanation for the impairment observed in our

experiment is that rats did not have sufficient time to make a choice during the acquisition. We

consider that this hypothesis should be further investigated. To do so, first, the TUNL task with

longer response intervals should be properly validated in WT rats, as such a modification can affect

the rats’ performance. Allowing for longer response intervals might confound working memory

results. Differently to the HO rats, HET rats were able to acquire the TUNL task at a similar rate

than the WT rat. Additionally, the behavioral results during the sessions with LFPs recordings in

the 2- and 8-seconds delays indicated that WT and HET rats were able to perform above chance

in the TUNL task.

Neuronal functioning was assessed by PSD and PAC. Our electrophysiological analysis

was centered on the TUNL task. One of the relevant characteristics of the TUNL task is the

possibility to compare the electrophysiological readouts during the sample and the choice stages.

Motorically, the rats are performing similar actions, but cognitively, they are executing distinct

processes, which allows investigating working memory processes without motoric confounders.

Our results indicate an increase in power for HET rats compared to WT, especially for the gamma

band at the mPFC without behavioral impairment. Previous research using the TgF344-AD rat

model also found power impairments in the mPFC, but rather than an increase in the power, an

attenuation of the gamma power was observed (Bazzigaluppi et al., 2018). It remains to be tested

if the PSD in the mPFC in the McGill-R-Thy1-APP rats decreases ones Aβ plaques are present.

Importantly, in the Bazzigalupi and colleagues study (Bazzigaluppi et al., 2018),

electrophysiological recordings were obtained while rats were under anesthesia. In our study, rats

were performing a behavioral task, which it is known to require the contribution of the mPFC for

successful working memory functioning (Benchenane, Tiesinga, & Battaglia, 2011). Contrary to

what was expected, we did not observe PAC differences between WT and HET rats for any of the

investigated brain areas at the tested age. It remains to be investigated if other brain areas or later

time points might reveal changes in PAC. Furthermore, we did not observe PSD and PAC changes

between the two stages of the task (sample and choice). The TUNL task allows the use of variable

delays, manipulating task complexity. Giving that the number of incorrect responses was very low

in our experiment, we only analyzed LFPs during correct responses. Longer delays could allow


P a g e 70 | 128

investigating differences in electrophysiological readouts between correct and incorrect responses,

or between sample and choice stages. Unfortunately, the low number of correct trials for the HO

rats did not allow us to record LFPs during the TUNL task. A second electrophysiological analysis

was conducted during home-cage environment exploration. In this part of the experiment, HO rats

were also included, as their performance in the TUNL task was not relevant for the analysis. Our

results indicated no difference in either power or coupling between the genotypes for any of the

recorded brain areas.

The selection of the 3.5 months of age aimed to investigate the early stage of Aβ pathology

without clear cognitive symptoms, and to match what is considered the preclinical phase of AD

(Sasaguri et al., 2017). It has been previously reported that McGill-R-Thy1-APP rats have

intracellular Aβ as early as 1 week of age and that this pathology is strongly established by 2 to 3

months of age in both HO and HET rats (Iulita et al., 2014, 2017; Leon et al., 2010; Wilson et al.,

2016). Importantly, our biochemical data indicate that the antibody, McSA1, used in the mentioned

studies (Iulita et al., 2014, 2017; Leon et al., 2010; Wilson et al., 2016) to identify Aβ pathology

in the McGill-R-Thy1-APP model, reacts to both sAPPα and Aβ 1-42. This implies that positive

immunohistochemical staining by this antibody is not clearly attributable to Aβ pathology but may

be caused by APP overexpression. Overall, contrary to current literature, our

immunohistochemical results with multiple antibodies indicate that the McGill-R-Thy1-APP rats

do not have intracellular Aβ pathology at the age investigated in this study. This leads us to the

conclude that any behavioral and electrophysiological deficits observed in our experiments are

likely associated with the overexpression of APP, rather than with Aβ amyloidosis.

AppNL-G-F Mice Characterization

P a g e 71 | 128

Behavioral, Electrophysiological and Histopathological Characterization of

AppNL-G-F Mice

Part of this chapter has been published in:

Jacob, S., Davies, G., De Bock, M., Hermans, B., Wintmolders, C., Bottelbergs, A., Borgers, M.,

Theunis, C., Van Broeck B., Manyakov, N. V., Balschun, D., Drinkenburg W.H.I.M. (2019).

Neural oscillations during cognitive processes in an App knock-in mouse model of Alzheimer’s

disease pathology. Scientific reports, 9 (1), 1-19.

5.1. Introduction

Numerous animal models mirroring some features of AD pathogenesis have been created

to facilitate the understanding of the molecular mechanisms of the disease (Jankowsky & Zheng,

2017). Among the most common models are transgenic mouse models overexpressing the human

APP with mutations linked to familial AD in APP or/and PSEN1/2 (Sasaguri et al., 2017).

Although these animals have been instrumental in understanding some basic aspects of AD

(Sasaguri et al., 2017), the overexpression of these proteins might lead to artificial phenotypes

(Born et al., 2014; Nilsson et al., 2014). At the brain network level, it has been reported by

numerous studies that these mouse models demonstrate various alterations (Busche & Konnerth,

2015; Palop et al., 2007; Palop & Mucke, 2010), even before Aβ accumulation (Goutagny et al.,

2013). Although these results show some phenotypical similarities with AD (Nimmrich et al.,

2015; Palop & Mucke, 2009; Poza et al., 2017), it is not clear what are the underlying mechanisms

that cause the aberrant neuronal activity observed in these models. Some evidence suggests that

APP overexpression, and not Aβ overproduction might be responsible for the abnormal network

activity in APP-overexpressing mouse models (Born et al., 2014). Further evidence supporting this

hypothesis, comes from studies investigating the physiological role of APP or its processed

fragments in non-pathological conditions (Maya & Bassem, 2014; Rice et al., 2019). For instance,

it has been demonstrated that sAPP plays a role in the modulation of synaptic transmission (Rice

et al., 2019).

The new generation of App KI mice produce robust Aβ amyloidosis with physiological

App levels (Nilsson et al., 2014; Saito et al., 2014a). While Aβ plaque deposition starts at 6 months


P a g e 72 | 128

in AppNL-F mice (Saito et al., 2014a), its onset in AppNL-G-F mice is already at 2 months and saturates

around 7 months (Saito et al., 2014a; Sakakibara, Sekiya, Saito, Saido, & Iijima, 2018). In contrast,

AppNL mice do not develop any Aβ plaques, even at late ages (Saito et al., 2014a; Sakakibara et

al., 2018). Behaviorally, AppNL-G-F mice demonstrate impairments in multiple cognitive domains,

including declined spatial reversal learning and attention control, loss of avoidance memory, and

enhanced impulsivity and compulsivity, starting at 8 months of age (Masuda et al., 2016). Studies

investigating early time points show more inconclusive results. For instance, while in the original

report, deficits in the Y-maze were indicated (Saito et al., 2014a), a later study did not find deficits

in working memory using the Y-maze at 6 months of age (Whyte et al., 2018). Another study

testing AppNL-G-F mice at 3, 6, and 10 months demonstrated largely unaffected behavioral readouts,

with only mildly changes in social and anxiety-related test performance (Latif-Hernandez et al.,

2019). Brown et al., investigated network hyperexcitability in two mouse models: AppNL-F and J20

(APP overexpression). Although they were able to replicate the previously reported effect on

network hyperexcitability in the J20 mice, the effect was absent in 𝐴𝑃𝑃𝑁𝐿−𝐹/𝑁𝐿−𝐹 mice (Brown

et al., 2018).

The App KI models have not yet been fully characterized with respect to neuronal network

activity. Therefore, the primary goal of this part of the thesis was to investigate

electrophysiological readouts in combination with a cognitive task to illuminate functional

differences between AppNL-G-F and C57BL/6J WT controls. Mice had to perform a VD task using

touchscreen operant boxes, starting at 4.5 months of age while LFPs in the DMS, Cg, RSC, and

dCA1 region of the hippocampus were recorded using a wireless neurophysiological signal

acquisition system. In parallel with these recordings, we measured LFPs at two time points during

home-cage exploration, without the behavioral task to investigate neuronal network changes

exclusively associated with pathology progression. The ages to perform the in vivo experiments

were selected with the objective to match the preclinical AD phase. Biochemical and

immunohistochemical analyses were conducted to correlate Aβ plaque deposition with the

electrophysiological and behavioral results at the different time points. The electrophysiological

analysis focused on PAC between theta (4 – 12 Hz), gamma (30 – 100 Hz), and HFO (100 – 200

Hz) activity during three distinct recording conditions related to different cognitive load, the start

and the end of the VD task, and exploration of the home environment.


P a g e 73 | 128

5.2. Results

Performance during VD Task

Pre-training:

During the pre-training stage there were no genotype effects on the number of sessions required in

the tone association (WT: M = 3.9, SD = 0.7, n = 10; AppNL-G-F: M = 4.5, SD = 0.5, n = 8, Z = 3.74,

p = 0.05), touch association (WT: M = 1.2, SD = 0.6, n = 10; AppNL-G-F: M = 1.0, SD = 0.0, n = 8,

Z = 0.8, p = 0.37), must touch (WT: M = 1.4, SD = 0.8, n = 10; AppNL-G-F: M = 1.4, SD = 1.1, n =

8, Z = 0.07, p = 0.78) and Punish Incorrect (WT: M = 3.9, SD = 1.4, n = 10; AppNL-G-F: M = 4.3,

SD = 1.4, n = 7, Z = 0.31, p = 0.57) stages of touchscreen pre-training. One AppNL-G-F mouse did

not complete all pre-training stages within 30 days and did not progress onto the VD task.

Visual Discrimination Task:

The individual learning curves of mice during acquisition of the VD task are shown in Figure 23,

a. As expected, percentage of correct responses on the first day of the discrimination task was

around chance level (WT: M = 49.17%, SD = 12.46, n = 10; AppNL-G-F: M = 52.97%, SD = 10.61,

n = 7) and no difference was found between genotypes t (15) = 0.656, p = 0.52. To compare

learning rates, a mixed linear model was used to fit of the percentage of correct responses data

(𝑅2= 0.715). There was a significant effect of session F (1, 139.6) = 252.3, p < 0.0001, and a

significant interaction session and genotype F (1, 139.6) = 29.07, p < 0.0001. The main genotype

effect did not reach significance F (1, 14.7) = 2.375, p = 0.1445. One mouse in the AppNL-G-F group

needed 24 sessions to reach the VD criterion, twice that of the slowest learner in the WT group.

Excluding this extreme value from the model fit resulted in a non-significant interaction term

between genotype and session F (1, 115.5) = 3.668, p = 0.0580. Analysis of the number of sessions

required to reach the learning criterion of two consecutive sessions of 80% correct responses or

higher did not identify a significant difference between groups (WT: M = 7.4%, SD = 2.91, n = 10;

AppNL-G-F: M = 11.43%, SD = 5.94, n = 7, Z= 2.95, p = 0.08, Figure 23, b). Similarly, a comparison

of the percentage of CTs did not indicate a genotype difference (WT: M = 32.24%, SD = 5.24, n =

10; AppNL-G-F: M = 33.54%, SD = 6.08, n = 7, Z = 0.15, p = 0.69, Figure 23, c). Analysis of latencies

to initiate trials (WT: M = 4.54%, SD = 1.86, n = 10; AppNL-G-F: M = 5.51%, SD = 2.9, n = 7, Z =

0.61, p =0.43, Figure 23, d), response to the stimulus (WT: M = 5.07%, SD =0.87, n = 10; AppNL-


P a g e 74 | 128

G-F: M = 7.69%, SD = 6.07, n = 7, Z = 0.95, p = 0.33, Figure 23, e), and to collect the food reward

(WT: M = 2.04%, SD = 0.34, n = 10; AppNL-G-F: M = 2.28%, SD = 0.89, n = 7, Z = 0.15, p = 0.69,

Figure 23, f) indicated no alterations in any response latencies for the AppNL-G-F mice. Together

these results suggest that the ability to discriminate visual stimuli and to form stimulus-reward

associations was largely unaffected in the AppNL-G-F mice.

Figure 23. Acquisition of the visual discrimination (VD) touchscreen task in AppNL-G-F and WT

mice. a. learning curves of mice during discrimination learning for WT (left panel) and AppNL-G-F

(right panel). Percentage of correct responses is plotted by sessions where each colored line

represents an individual mouse. b. number of sessions to achieve learning criterion of two

consecutive sessions of 80% correct or higher responses did not differ between WT and AppNL-G-F

mice. c. percentage of correction trials did not differ between genotypes. Latencies to initiate trial

(d), response to the stimulus (e), and collect the reward (f) did not differ between genotypes. In

figures b, c, d, e, f data are represented in box plots with individual points for each subject.

Statistically significance level at p < 0.05. ns: non-significant.


P a g e 75 | 128

Brain Oscillations Analysis During VD Task

Relative Power Spectral Density:

Before investigating the relative PSD in the 4 brain areas of interest during the start and end of the

VD task, we grouped frequencies on three bands: theta (4 to 12 Hz), gamma (30 to 100 Hz) and

HFO (101 to 200 Hz), based on previous proposed guidelines(Jobert et al., 2012). A within-subject

analysis between the end and start of the VD task revealed no significant difference in relative

PSD for any of the frequency bands for both genotypes (see Figure 24 with Cg brain area as an

example and Table 12 for all descriptive statistics and statistical tests). We also investigated

relative PSD between WT and AppNL-G-F, but no significant differences were found for any of the

brain areas and frequency bands.

Figure 24. Mean relative power spectral density (PSD) across animals during the visual

discrimination (VD) task for the cingulate cortex (Cg). Mean relative PSD in the 4 to 200 Hz band

during start of the VD task (Task_Start, black and blue curves) and end of the VD task (Task_End,

green and red curves) for (a) WT and (c) AppNL-G-F mice. Delta between Task_End and Task_Start

is represented in box plots were individual points indicate delta values for each subject for (b) WT

and (d) AppNL-G-F mice. No significant difference was observed between the end and the start of

the task for the two genotypes at the three different frequency bands. Significance level at q = 0.05,

false discovery rate (FDR) corrected for multiple comparisons. ns: non-significant.


P a g e 76 | 128

Table 12. Relative power spectral density (PSD, µV2/Hz) during the visual discrimination (VD)

task.

Brain areas: Cingulate Cortex (Cg), dorsal CA1 region of the hippocampus (dCA1), dorsal medial

striatum (DMS), and retrosplenial cortex (RSC).

Frequencies: theta, gamma and high frequency oscillations (HFO).

Conditions: Task_Start = start of the VD task, Task_End = end of the VD task.

Descriptive statistics for PSD (µV2/Hz), N: sample size, Z: Wilcoxon rank-sum test.


Genotype Brain Area Freq Condition Mean SD Median Range N Z Prob > [Z]

p value

rank

FDR

threshold

Task_Start 0,03879 0,01048 0,04082 0,03418

Task_End 0,03639 0,00094 0,03746 0,02788

Task_Start 0,00540 0,00110 0,00510 0,00360

Task_End 0,00560 0,00100 0,00550 0,00300

Task_Start 0,00120 0,00034 0,00120 0,00078

Task_End 0,00130 0,00036 0,00120 0,00082

Task_Start 0,04875 0,00666 0,04702 0,01918

Task_End 0,04600 0,00860 0,04278 0,02300

Task_Start 0,00450 0,00077 0,00460 0,00220

Task_End 0,00480 0,00083 0,00530 0,00220

Task_Start 0,00093 0,00020 0,00100 0,00049

Task_End 0,00099 0,00022 0,00100 0,00059

Task_Start 0,04992 0,01000 0,05291 0,03031

Task_End 0,04901 0,00896 0,05110 0,02459

Task_Start 0,00420 0,00062 0,00440 0,00200

Task_End 0,00440 0,00068 0,00430 0,00190

Task_Start 0,00019 0,00006 0,00020 0,00017

Task_End 0,00025 0,00013 0,00023 0,00049

Task_Start 0,05863 0,00790 0,06198 0,01952

Task_End 0,05361 0,00862 0,05439 0,02604

Task_Start 0,00360 0,00100 0,00330 0,00300

Task_End 0,00410 0,00120 0,00380 0,00380

Task_Start 0,00026 0,00011 0,00026 0,00031

Task_End 0,00031 0,00013 0,00032 0,00038

Task_Start 0,06310 0,00684 0,06254 0,02294

Task_End 0,06634 0,00559 0,06502 0,01872

Task_Start 0,00320 0,00076 0,00300 0,00250

Task_End 0,00330 0,00064 0,00340 0,00220

Task_Start 0,00042 0,00013 0,00037 0,00043

Task_End 0,00041 0,00012 0,00039 0,00037

Task_Start 0,06536 0,00494 0,06540 0,01470

Task_End 0,06330 0,01026 0,06251 0,02893

Task_Start 0,00320 0,00069 0,00310 0,00210

Task_End 0,00360 0,00098 0,00320 0,00250

Task_Start 0,00040 0,00014 0,00039 0,00041

Task_End 0,00041 0,00014 0,00036 0,00041

Task_Start 0,04116 0,00835 0,03904 0,02734

Task_End 0,03915 0,00799 0,03650 0,02338

Task_Start 0,00480 0,00088 0,00490 0,00300

Task_End 0,00510 0,00091 0,00510 0,00280

Task_Start 0,00119 0,00033 0,00125 0,00093

Task_End 0,00132 0,00031 0,00148 0,00071

Task_Start 0,04553 0,01188 0,04353 0,03560

Task_End 0,04530 0,01254 0,04611 0,03356

Task_Start 0,00460 0,00120 0,00460 0,00370

Task_End 0,00470 0,00120 0,00480 0,00330

Task_Start 0,00109 0,00027 0,00098 0,00073

Task_End 0,00103 0,00028 0,00110 0,00078

Gamma

HFO

Cg

dCA1

DMS

RSC

WT

App NL-G-F

WT

App NL-G-F

WT

App NL-G-F

WT

App NL-G-F

HFO

HFO

Theta

0,00833

6

7 12 0,0469 4

Theta

Gamma

HFO

Theta

Gamma

HFO

Theta

Gamma

HFO

Theta

Gamma

10 0,1953 16 0,03333

10

8

0,04375

9 0,1563 14 0,02917

7

7

24 0,05000

0,01042

7 0,01458

4,5 0,6953 22 0,04583

0,2969 18 0,03750

Theta

Gamma

HFO

Theta

Gamma

HFO

Theta

Gamma

9,5

0,03125

8 13 0,0781 9

0,01250

7 7 0,2969 17 0,03542

11 0,0781 8 0,01667

6 9,5 0,0625 6

7

7

8

9

0,0625

0,1484 13

8

5 0,5469

0,0156

8 -11

9 0,1563 15

5

3 0,6875 21

0,02708

0,01875

10 25,5 0,0059 1 0,00208

20 0,04167

15,5 0,0742

9 -13,5 0,1289 12

0,04792

8 17

7 -1 0,9375

2 0,00417

10 -1,5 0,9219 23

0,006259 19,5 0,0195

0,02500

7 -10 0,1094 11 0,02292

3

-7 0,3828 19 0,03958

0,02083

7 -7

6 -8,5 0,0938 10


P a g e 77 | 128

Phase-Amplitude Coupling:

Similarly to relative PSD analysis, we investigated whether the MI changed as the animals

progressed on the VD task and if this modulation was different for the AppNL-G-F compared to the

WT mice. For this analysis, we grouped amplitude frequencies in three bands: LG (30 - 60Hz),

HG (61 - 100 Hz), and HFO (101 - 200 Hz). This banding was selected as it has been suggested

that different frequency bands act as different channels for communication. For instance, in the

CA1 region of the hippocampus, gamma oscillations split into two different components indicating

different origins. LG arises from interactions with the CA3 region of the hippocampus, while HG

is thought to arise from interactions with the entorhinal cortex (Colgin et al., 2009). A visual

inspection of the phase-amplitude comodulograms of the different brain areas indicates that PAC

occurs in different frequency bandings. For instance, Cg (Figure 25, a) and RSC (Figure 26, a)

show coupling between HFO and higher theta, while the DMS (Figure 27, a) and dCA1 (Figure

28, a) show coupling between HG and high theta. When investigating the changes in the MI

between the end and the start of the VD task the most prominent effect was observed in the Cg

(Figure 25, b) and RSC (Figure 26, b) for WT mice where the MI decreased by the end of the VD.

WT mice also demonstrated a similar effect in HFO for the DMS (Figure 27, b) and dCA1 (Figure

28, b). Importantly, during the start of the task, the coupling seems to be associated with more

frequency bands, suggesting that as the mice learned the task, the coupling became more localized.

These differences were not clearly observed for the AppNL-G-F mice, where they only demonstrated

a decreased coupling for HG in the Cg (Figure 25, b). See Table 13, for completed descriptive

statistics and results of statistical analysis. Next, we investigated genotype differences within each

condition (task-start and task-end). For this, we compared the MI for the three different amplitude

bands at the start and the end of the task and corrected for multiple comparisons using a FDR, as

done with all previous comparisons (see 3.9 Statistical Analysis). Table 14 shows that no

significant differences were observed between AppNL-G-F and WT mice for any of the brain areas

and frequency bands. Note that for simplicity, panel c of figures: Figure 25, Figure 26, Figure 27,

and Figure 28 shows box plots of the full amplitude range, but the statistical analysis was done for

each separate frequency band. It should be noted that some of the mean phase-amplitude

comodulograms might visually suggest changes between genotypes, but as it can be seen in the

box plots, some of these measurements had a high level of variability. The different results


P a g e 78 | 128

obtained for the within and between-subject analysis might reflect difference in the experimental

design. Importantly, these designs are addressing different questions, therefore, they may provide

different patterns of results (Keren, 2014).

Figure 25. Mean phase-amplitude coupling (PAC) during the visual discrimination (VD) task for

the cingulate cortex (Cg). a. Phase-amplitude comodulograms plotted for WT mice (top panels)

and AppNL-G-F mice (lower panels) for the start of the VD task (Task_Start, left) and the end of the

VD task (Task_End, right). b. Modulation index (MI) delta between Task_End and Task_Start for

low gamma (LG, left), high gamma (HG, middle) and high frequency oscillations (HFO, right) for

WT mice (top panels) and AppNL-G-F mice (lower panels). Comparison at each frequency band

showed a significant difference for WT mice, but only at HG for AppNL-G-F mice. c. MI between

WT and AppNL-G-F mice for Task_Start (left) and Task_End (right). Comparison at each frequency

band showed no significant difference. Values are represented in box plots were individual points

indicate (b) delta or (c) absolute values for each subject. Significance level at q = 0.05, false

discovery rate (FDR) corrected for multiple comparisons. ns: non-significant.


P a g e 79 | 128


the retrosplenial cortex (RSC). a. Phase-amplitude comodulograms plotted for WT mice (top

panels) and AppNL-G-F mice (lower panels) for the start of the VD task (Task_Start, left) and the

end of the VD task (Task_End, right). b. Modulation index (MI) delta between Task_End and

Task_Start for low gamma (LG, left), high gamma (HG, middle) and high frequency oscillations

(HFO, right) for WT mice (top panels) and AppNL-G-F mice (lower panels). Comparison at each

frequency band showed a significant difference for WT mice at LG and HG, but no significant

difference for AppNL-G-F mice. c. MI between WT and AppNL-G-F mice for Task_Start (left) and

Task_End (right). Comparison at each frequency band showed no significant difference. Values

are represented in box plots were individual points indicate (b) delta or (c) absolute values for each

subject. Significance level at q = 0.05, false discovery rate (FDR) corrected for multiple

comparisons. ns: non-significant.


P a g e 80 | 128


the dorsal medial striatum (DMS). a. Phase-amplitude comodulograms plotted for WT mice (top

panels) and AppNL-G-F mice (lower panels) for the start of the VD task (Task_Start, left) and the

end of the VD task (Task_End, right). b. Modulation index (MI) delta between Task_End and

Task_Start for low gamma (LG, left), high gamma (HG, middle) and high frequency oscillations

(HFO, right) for WT mice (top panels) and AppNL-G-F mice (lower panels). Comparison at each

frequency band showed a significant difference for WT mice at HFO, but no significant difference

for AppNL-G-F mice. c. MI between WT and AppNL-G-F mice for Task_Start (left) and Task_End

(right). Comparison at each frequency band showed no significant difference. Values are

represented in box plots were individual points indicate (b) delta or (c) absolute values for each

subject. Significance level at q = 0.05, false discovery rate (FDR) corrected for multiple



P a g e 81 | 128


the dorsal CA1 region of the hippocampus (dCA1). a. Phase-amplitude comodulograms plotted

for WT mice (top panels) and AppNL-G-F mice (lower panels) for the start of the VD task

(Task_Start, left) and the end of the VD task (Task_End, right). b. Modulation index (MI) delta

between Task_End and Task_Start for low gamma (LG, left), high gamma (HG, middle) and high

frequency oscillations (HFO, right) for WT mice (top panels) and AppNL-G-F mice (lower panels).

Comparison at each frequency band showed a significant difference for WT mice at HFO, but no

significant difference for AppNL-G-F mice. c. MI between WT and AppNL-G-F mice for Task_Start

(left) and Task_End (right). Comparison at each frequency band showed no significant difference.

Values are represented in box plots were individual points indicate (b) delta or (c) absolute values

for each subject. Significance level at q = 0.05, false discovery rate (FDR) corrected for multiple



P a g e 82 | 128

Table 13. Phase-amplitude coupling (PAC) during the visual discrimination (VD) task.



Frequencies: theta, gamma and high frequency oscillations (HFO).


Descriptive statistics for Modulation Index (MI), N: sample size, Z: Wilcoxon rank-sum test.


Grey cells indicated significance values. P value ranks lower than 8 are statistically significant

different between the two conditions.

Genotype Brain Area Freq Condition Mean SD Median Range N Z Prob > [Z]

p value

rank

FDR

threshold

Task_Start 0,000136 0,000054 0,000133 0,000150

Task_End 0,000060 0,000020 0,000066 0,000048

Task_Start 0,000167 0,000056 0,000164 0,000201

Task_End 0,000072 0,000034 0,000065 0,000094

Task_Start 0,000271 0,000060 0,000283 0,000193

Task_End 0,000138 0,000051 0,000143 0,000172

Task_Start 0,000119 0,000074 0,000124 0,000214

Task_End 0,000073 0,000035 0,000085 0,000084

Task_Start 0,000219 0,000173 0,000181 0,000506

Task_End 0,000099 0,000040 0,000098 0,000120

Task_Start 0,000247 0,000090 0,000229 0,000249

Task_End 0,000135 0,000065 0,000099 0,000174

Task_Start 0,000852 0,000477 0,000713 0,001435

Task_End 0,000719 0,000499 0,000732 0,001614

Task_Start 0,003142 0,003040 0,001842 0,002389

Task_End 0,002817 0,003087 0,001674 0,008305

Task_Start 0,001266 0,001129 0,000807 0,003486

Task_End 0,001019 0,001000 0,000786 0,003212

Task_Start 0,000332 0,000204 0,000310 0,000581

Task_End 0,000371 0,000320 0,000257 0,000816

Task_Start 0,001877 0,001494 0,001523 0,004300

Task_End 0,001438 0,008713 0,001322 0,002162

Task_Start 0,000597 0,000397 0,000438 0,001036

Task_End 0,000398 0,000182 0,000377 0,000465

Task_Start 0,000176 0,000071 0,000156 0,000201

Task_End 0,000115 0,000036 0,000119 0,000092

Task_Start 0,000434 0,000149 0,000443 0,000412

Task_End 0,000322 0,000130 0,000318 0,000340

Task_Start 0,000434 0,000112 0,000433 0,000326

Task_End 0,000276 0,000121 0,000296 0,000390

Task_Start 0,000190 0,000074 0,000205 0,000205

Task_End 0,000136 0,000047 0,000158 0,000120

Task_Start 0,000703 0,000571 0,000494 0,001642

Task_End 0,000535 0,000505 0,000267 0,001350

Task_Start 0,000350 0,000098 0,000360 0,000246

Task_End 0,000230 0,000154 0,000188 0,000480

Task_Start 0,000147 0,000071 0,000119 0,000218

Task_End 0,000075 0,000023 0,000075 0,000067

Task_Start 0,000177 0,000053 0,000181 0,000189

Task_End 0,000091 0,000034 0,000087 0,000095

Task_Start 0,000309 0,000097 0,000326 0,000332

Task_End 0,000196 0,000068 0,000182 0,000178

Task_Start 0,000130 0,000091 0,000089 0,000268

Task_End 0,000084 0,000043 0,000084 0,000129

Task_Start 0,000202 0,000118 0,000188 0,000336

Task_End 0,000098 0,000044 0,000089 0,000120

Task_Start 0,000272 0,000089 0,000302 0,000225

Task_End 0,000166 0,000096 0,000155 0,0002890,0229167

0,0354167

0,0291667

0,0270833

0,01875

0,00625

0,0145833

0,0125

0,0020833

0,0416667

0,03125

3

2

21

16

0,0479167

0,0458333

0,025

0,0208333

0,0104167

0,0166667

0,0395833

0,0375

0,0083333

0,05

0,0041667

0,04375

0,0333333

11

5

17

14

13

9

0,1094

0,0313

7

6

1

20

15

8

19

18

4

24

23

22

12

10

0,0781

0,0195

0,0078

0,0078

0,375

0,0391

0,0234

0,0156

0,1563

0,0781

-21,5

-6

-17

-17

-18

-8

-11

-10

-13

0,0156

0,0156

0,0078

0,2188

0,0781

0,0156

0,1934

0,1602

0,0098

0,8438

0,5625

0,4375

-9

-11

-11

-19,5

-21,5

8

7

9

7

HFO

LG

HG

HFO

-14

-13,5

-14,5

-24,5

1,5

-3,5

-4,5

-15

-16

-17

8

7

10

LG

HG

HFO

LG

HG

HG

HFO

LG

HG

HFO

HFO

LG

HG

HFO

LG

LG

HG

HFO

LG

HG

6

WT

RSC

App NL-G-F

WT

Cg

App NL-G-F

WT

dCA1

App NL-G-F

WT

DMS

App NL-G-F


P a g e 83 | 128

Table 14. Between-subject analysis of phase-amplitude coupling (PAC) during the visual

discrimination (VD) task.




Frequencies: low gamma (LG), high gamma (HG), and high frequency oscillations (HFO).



Brain Oscillations Analysis During Home-cage Environment Exploration

Relative Power Spectral Density:

Our next analysis was conducted to determine if we could observe different brain oscillatory

activity when animals were exploring the home-cage environment at an age of 5 months compared

to 8 months. As with the relative PSD during the VD task, we did not observe difference for any

of the frequency bands between genotypes (Table 15). Furthermore, age did not have an effect of

the PSD for either the AppNL-G-F and WT mice.

Condition Brain Area Freq Z Prob > [Z] p value rank

FDR

threshold

LG -0,6365 0,5244 14 0,0292

HG 0,28932 0,7723 20 0,0417

HFO -0,6365 0,5244 15 0,0313

LG -2,22354 0,0262 1 0,0021

HG -0,27116 0,7863 21 0,0438

HFO -1,57275 0,1158 2 0,0042

LG 0,17359 0,8622 23 0,0479

HG 0,98368 0,3253 9 0,0188

HFO -1,33087 0,1832 5 0,0104

LG -0,74096 0,4587 12 0,0250

HG 0,10585 0,9157 24 0,0500

HFO -0,74096 0,4587 13 0,0271

LG 0,86796 0,3854 11 0,0229

HG 1,56232 0,1182 3 0,0063

HFO -0,6365 0,5244 16 0,0333

LG -1,24735 0,2123 6 0,0125

HG -0,59656 0,5508 17 0,0354

HFO -1,24735 0,2123 7 0,0146

LG 1,33087 0,1832 4 0,0083

HG 0,40505 0,6854 19 0,0396

HFO -1,09941 0,2716 8 0,0167

LG 0,2117 0,8323 22 0,0458

HG 0,4234 0,672 18 0,0375

HFO -0,95266 0,3408 10 0,0208

Task_Start

Task_End

Cg

dCA1

DMS

RSC

Cg

dCA1

DMS

RSC


P a g e 84 | 128

Table 15. Between-subject analysis of relative power spectral density (PSD) during the no-task

condition. Conditions: 5 and 8 months of age.



Frequencies: theta, gamma, and high frequency oscillations (HFO).



Phase-amplitude Coupling:

As a with the relative PSD, no significant differences were observed between genotypes during

the home-cage environment exploration (Table 16) for any of the brain areas. Furthermore, we did

not observe an effect on age for either the WT or AppNL-G-F mice. Taken together, these results

suggest that the AppNL-G-F mice demonstrate up to 8 months of age normal PAC in a freely moving

condition without task performance (investigation of further age points were beyond the scope of

this study).

Condition Brain Area Freq Z Prob > [Z] p value rank

FDR

threshold

Theta 0,6365 0,5244 14 0,0291667

Gamma -0,05786 0,9539 24 0,05

HFO -0,40505 0,6854 19 0,0395833

Theta 0,88388 0,3768 9 0,01875

Gamma 0,29463 0,7683 20 0,0416667

HFO 0,76603 0,4437 11 0,0229167

Theta 0,63888 0,5229 13 0,0270833

Gamma -0,5111 0,6093 17 0,0354167

HFO -0,7665 0,4433 10 0,0208333

Theta 0,28932 0,7723 21 0,04375

Gamma -0,40505 0,6854 18 0,0375

HFO -0,52077 0,6025 16 0,0333333

Theta -0,96825 0,3329 7 0,0145833

Gamma 0,71005 0,4777 12 0,025

HFO 1,74284 0,0814 2 0,0041667

Theta -0,13333 0,8415 23 0,0479167

Gamma 0,26667 0,7897 22 0,0458333

HFO 1,2 0,2301 5 0,0104167

Theta -1,5 0,1336 3 0,00625

Gamma 2,07143 0,0383 1 0,0020833

HFO 0,92857 0,3531 8 0,0166667

Theta 1,09735 0,2725 6 0,0125

Gamma -0,58095 0,5613 15 0,03125

HFO -1,22644 0,22 4 0,0083333

5_Months

8_Months

Cg

dCA1

DMS

RSC

Cg

dCA1

DMS

RSC


P a g e 85 | 128

Table 16. Between-subject analysis of phase-amplitude coupling (PAC) during the no-task

condition.

Conditions: 5 and 8 months of age.



Frequencies: low gamma (LG), high gamma (HG), and high frequency oscillations (HFO).



Pathology

Biochemical analysis of AppNL-G-F forebrain tissue revealed that Aβ1-42 increased in an age-

dependent manner (Figure 29, a) and was accompanied by progressive Aβ plaque pathology as

seen by immunohistochemistry (Figure 29, b-c). Aβ1-42 levels in the brain seemed to be minor at

2 months of age, but progressively increased starting from 4 months of age up to 12-15 months.

This can also be deduced from the immunohistochemical evaluation of cortical and hippocampal

pathology (% Aβ plaque area as detected by antibodies JRF/cAβ42/26 and JRF/AβN/25) that

showed a limited number of Aβ deposits at 2 months of age in cortex, further progressing up to 8-

Condition Brain Area Freq Z Prob > [Z] p value rankFDR

threshold

LG -0,68264 0,4948 12 0,02609

HG 1,31276 0,1893 6 0,01304

HFO -0,26255 0,7929 22 0,04783

LG -1,37607 0,1688 3 0,00652

HG -0,52926 0,5966 14 0,03043

HFO -0,31755 0,7508 19 0,04130

LG 1,44659 0,148 1 0,00217

HG 0,52077 0,6025 15 0,03261

HFO 0,28932 0,7723 21 0,04565

LG 0,36757 0,7132 18 0,03913

HG 1,20774 0,2271 7 0,01522

HFO -0,15753 0,8748 24 0,05217

LG -0,40505 0,6854 16 0,03478

HG -0,40505 0,6854 17 0,03696

HFO 1,33087 0,1832 5 0,01087

LG -1,35529 0,1753 4 0,00870

HG -0,88388 0,3768 10 0,02174

HFO -0,53033 0,5959 13 0,02826

LG 0,7665 0,4433 11 0,02391

HG 1,14998 0,2502 8 0,01739

HFO 0,89443 0,3711 9 0,01957

LG -0,28932 0,7723 20 0,04348

HG -0,17359 0,8622 23 0,05000

HFO 1,44659 0,148 2 0,00435

5_Months

Cg

dCA1

DMS

RSC

8_Months

Cg

dCA1

DMS

RSC


P a g e 86 | 128

15 months as observed by others (Masuda et al., 2016; Saito et al., 2014a). Altogether, our data

indicate that between 4 and 8 months of age, the period where mice were used for the in vivo

behavioral and neurophysiological characterization, increased Aβ1-42 levels and Aβ plaques were

present in both cortex and hippocampus.

Figure 29. Amyloid β pathology in AppNL-G-F mice. a. Biochemical quantification of Aβ1-42 in

AppNL-G-F forebrains (GuHCl fraction) measured by sandwich ELISA. Standard curves for

calibration were generated using synthetic human Aβ1–42 peptide. Data represent the mean of 2

to 4 independent measurements of pooled samples with each pool consisting of equal volumes of

extracts from 4 to 6 mice (n = 5, 6, 5, 4, 4, 6, 6 per indicated time point respectively). Star symbols

indicate significance compared to 2 mo (months of age). b and c. Example images and

immunohistochemical analysis of the plaque occupancy area in cortex (inset) and hippocampus

using a C-terminal antibody (JRF/cAβ42/26, b) or N-terminal antibody (JRF/AβN/25, c). Scale

bars in the photographs are 2.5 mm and 50 µm (inset). In the summarizing plots, mean and SD are

presented in a bar chart in which each superimposed dot representing a single animal. Star symbols

indicate significance compared to 2 mo.


P a g e 87 | 128

5.3. Conclusion

The purpose of this part of the thesis was to characterize the onset and progression of AD-mediated

deficits in brain network functions in AppNL-G-F mice, a second-generation mouse model of Aβ

amyloidosis(Saito et al., 2014a), by behavioral, electrophysiological, and pathological

investigations in an age-dependent manner. Previous research using human APP-overexpressing

models has led to the hypothesis that aberrant network and altered oscillatory rhythmic activity

caused by Aβ amyloidosis reflect and underlie the early cognitive disturbances observed in AD

(Busche & Konnerth, 2015; Palop & Mucke, 2009, 2016). In our study, we tested this hypothesis

using a more relevant model of AD pathology, as the AppNL-G-F mice produce robust Aβ

amyloidosis, like the first-generation models (LaFerla & Green, 2012) and AD patients (Scheltens

et al., 2016), but without potential undesirable side effects caused by the overexpression of APP

(Born et al., 2014; Nilsson et al., 2014).

Behaviorally, the AppNL-G-F mice did not show strong associative learning impairments in

the VD task. Overall AppNL-G-F mice demonstrated a higher variability than the controls in learning

the task. Mice were 4.5 months old at the start of the VD task, which correlates with early plaque

deposition. Previous studies characterizing the new generation of AD mouse models have

identified some minimal behavioral changes at young ages (Latif-Hernandez et al., 2019), though

impaired performance on cognitive tests has only been reported from the age of 6 months onwards

(Masuda et al., 2016; Saito et al., 2014a; Whyte et al., 2018). Our study, using a different

behavioral task than those already investigated in this model, contributes to an increasing body of

evidence indicating that the AppNL-G-F mice do not show severe cognitive impairments at early

plaque stages.

To investigate the electrophysiological correlates, we used advanced techniques to explore

the effects of AD-related pathology on neural oscillations in different brain areas during cognition.

Our analysis was centered on comparing the start and end of the VD task. At the end of the task,

the relationship between decision making and coupling could be easily assessed, as mice were at

80% or higher accuracy. For the start of the task, the interpretation was more difficult, as mice

were responding at chance level. Therefore, some of the correct trial response analysis could

include brain activity that may or may not pertain to decision making. We believe that some of the

non-specific coupling observed at the start of the task could be associated with this mentioned


P a g e 88 | 128

variability within correct responses. Our results indicate a decrease in coupling for the WT mice

as they learned the task, especially for the Cg and RSC. This result is in accordance with previous

research indicating that PAC increases with task difficulty (Axmacher et al., 2010; Tamura,

Spellman, Rosen, Gogos, & Gordon, 2017). For the AppNL-G-F mice, this decrease was less

pronounced, only reaching statistical significance for HG in the Cg. Importantly, between-subject

comparisons for PAC and PSD did not reveal significant differences between the two genotypes.

Altogether our results indicate very subtle changes in coupling, suggesting PAC might not be

affected in the preclinical phase of AD based on amyloidosis pathology only. It remains to be

tested if other brain areas or the interaction between them might reveal changes in PAC at these

early stages. Furthermore, other complex functional connectivity analysis that has been explored

in patients (Engels et al., 2015), should be investigated in this and other animal models of AD

pathology.

Our main motivation for this study was to investigate PAC, for which a role in neuronal

information processing was reported (Lega, Burke, Jacobs, & Kahana, 2016), in a second-

generation mouse model while animals performed the task. In addition, we were also interested in

investigating PAC without any influence that the VD task could have on this readout. To this end,

we measured LFPs when mice were exploring their home-cage environment at 5 and 8 months of

age. Despite seeing an age-dependent increase in Aβ pathology between these time points, we did

not observe changes in PAC for any of the brain regions examined. Importantly, the ages for

recording were selected based on previous reports of pathology (Saito et al., 2014a). Our objective

was to obtain two measurements: one at a relatively early plaque stage and a second one at a later

time point where amyloid pathology is at an advanced state. Our findings indicate an age-

dependent increase in Aβ1-42 levels and plaque deposition that are consistent with previous

research (Masuda et al., 2016; Mehla et al., 2019; Saito et al., 2014a).

Altogether, our results from this study do not support the hypothesis of early alterations in

oscillations and functional neuronal activity (Busche & Konnerth, 2015; Goutagny et al., 2013;

Palop & Mucke, 2016) postulated using the first-generation models. We believe that by allowing

researchers to dissociate the effects of APP overexpression from the pathophysiological changes

in AD, the second-generation models will facilitate a deeper understanding of the neurobiology of

the disease.

General Discussion

P a g e 89 | 128

General Discussion

In this thesis, we characterized the neural circuitry dynamics during cognitive processes in two

experimental rodent models of AD pathology, the AppNL-G-F mouse model and the McGill-R-Thy1-

APP rat model. Evidence suggests that the pathological processes underlying AD start decades

prior to the manifestation of cognitive symptoms (Dubois et al., 2016). We hypothesized that Aβ

amyloidosis causes aberrant network activity at the preclinical stage of the disease, which reflects

the early cognitive disturbances observed in models of AD pathology.

As an initial step in this project, we were able to successfully implement state-of-the-art

neurophysiological imaging techniques during cognitive processes, combining operant

touchscreen apparatus with wireless LFP recordings. This approach was selected because of

multiple advantages: firstly, the use of a wireless system allowed the rodents to move freely in the

operant box. Secondly, the touchscreen platform permitted standardization and lowered the

motoric demands on the rodents (Horner et al., 2013). Finally, the synchronization of LFPs with

different behavioral parameters enabled a precise quantification of electrophysiological changes

related to behavioral performance; thus, making optimal use of the high temporal resolution of

LFPs recording techniques. We also implemented, optimized, and validated surgical implantation

of electrodes for the LFP recordings.

Electrode placement was motivated by two factors. The first factor was based on

topographic distribution of amyloid plaques in AD patients (Serrano-Pozo et al., 2011) and animal

models with Aβ pathology (Sasaguri et al., 2017). Following the Braak and Braak

neuropathological staging of AD-related changes, amyloid deposits are first observed in the

neocortex and gradually extend to the hippocampal formation at the latest stages of the disease

(Braak & Braak, 1991). Importantly, the distribution of amyloid deposits exhibits considerable

inter-individual variability (Braak & Braak, 1991). Of note, in carriers of the Arctic mutation, sub-

cortical pathology is observed (Kalimo et al., 2013). Aβ pathology distribution varies among the

different animal models. The AppNL-G-F mice have been reported to show cortical amyloidosis

starting at 2 months of age extending to sub-cortical areas by 4 months, recapitulating patients

carrying the Arctic mutation (Saito et al., 2014b). These results have been confirmed by our

immunohistochemical studies. The McGill-R-Thy1-APP rats have been reported to have abundant

intraneuronal accumulation of Aβ in the cortex and hippocampus by 3 months of age. As will

General Discussion

P a g e 90 | 128

further be discussed in this discussion, this result was not confirmed in this thesis. The second

factor to motivate our electrode placement was based on the functional role of these areas on

cognitive processes associated with the two behavioral tasks (Benchenane et al., 2011; Thorn,

Atallah, Howe, & Graybiel, 2010). Highly relevant areas that are associated with the used tasks

are the DMS and the mPFC because they play a role in establishing action-reward associations in

the VD task (Thorn et al., 2010), and working memory function in the TUNL task respectively

(Benchenane et al., 2011).

Part of our objectives was to study and characterize oscillations during cognitive processes

in models of AD pathology at an age that matches the preclinical phase of AD. To this end, we

characterized of the McGill-R-Thy1-APP rat model at a behavioral, electrophysiological and

histopathological level. We observed behavioral impairments in the HO rats, as these animals

failed to acquire the TUNL task. At the electrophysiological level, we observed an increase in PSD

in HET rats compared to WT during the task. Importantly and in contrast to what was expected

based on existing literature (Leon et al., 2010), we did not observe any Aβ pathology in both HO

and HET rats at the age investigated in this study. This suggests that the observed changes in our

experiments are likely associated with the artificial overexpression of APP. The second rodent

model used in this thesis was the AppNL-G-F mouse model. As with the rats, we characterized these

mice at behavioral, electrophysiological and histopathological level. Our findings confirmed an

age-dependent increase in Aβ1-42 levels and plaque deposition in these mice. This had no

consequences on cognitive performance in the VD task, which was largely unaffected in the AppNL-

G-F mice at the ages tested. At electrophysiological level, we observed subtle changes in PAC in

the WT mice during the VD task that were not present in the AppNL-G-F mice. Furthermore, we did

not detect age-dependent changes in PSD and PAC for both the WT and the AppNL-G-F mice. Our

results obtained in this mouse model with physiological App expression levels and high plaque

load do not provide support for the hypothesis that aberrant network activity, caused by Aβ

amyloidosis, could be used as a potential readout to detect any functional or cognitive alterations

in early AD pathology.

Although validating transgenic models was not part of the initial scope of this thesis,

our results highlight the importance of selecting and validating relevant animal models to obtain

meaningful results. Numerous publications using the McGill-R-Thy1-APP rat model have

suggested the existence of intraneuronal Aβ species. The observation of functional impairments

General Discussion

P a g e 91 | 128

prior to the formation of any extracellular plaques led to the hypothesis that intraneuronal

accumulation of Aβ is a toxic event driving AD (Iulita et al., 2014; Leon et al., 2010; Qi et al.,

2014; Wilson et al., 2016). The presence of intraneuronal accumulations of Aβ was based on

immunohistochemistry using the McSA1 monoclonal antibody to human Aβ. To prove the

specificity of McSA1 for Aβ peptides over human full-length APP, these studies used antibody

preabsorption (Leon et al., 2010) and colocalization (Iulita et al., 2014) techniques. Following our

ELISA evaluations, our findings challenge this view and indicate that the observed effects in this

study are independent of intracellular Aβ accumulation, and seem merely a consequence of the

overexpression of APP. Although more studies should be conducted to confirm these findings, we

believe our ELISA results are compelling evidence that the McSA1 antibody cannot distinguish

Aβ peptides from the same Aβ epitopes within APP. This calls for a careful re-evaluation of any

reported findings in this animal model. Special attention should be paid to studies claiming

functional consequences of intracellular Aβ accumulation (Parent et al., 2017; Qi et al., 2014).

Similarly, reports evaluating intraneuronal Aβ in other APP overexpression animal models should

carefully determine antibody selectivity (Goutagny et al., 2013; Palop & Mucke, 2010).

Among the most commonly used models are transgenic rodent overexpressing human

APP with mutations linked to familial AD resulting in amyloidosis. Importantly, the normal

physiological function of APP is not completely understood. Evidence suggest that sAPP is a

modulator of synaptic transmission (Mockett, Richter, Abraham, & Müller, 2017). In a recent

study, Rice and colleagues, identified the ɣ-aminobutyric acid type B receptor (GABABR), as

the candidate receptor for sAPP mediating synaptic activity (Rice et al., 2019). In 2014, Born et

al., disseminated the effects of APP overexpression, Aβ overproduction, and plaque formation on

network activity. Their results indicate that APP overexpression, and not Aβ pathology, is

responsible for neuronal network alterations observed in APP-overexpressing mouse models (Born

et al., 2014). Taken together, the mentioned publications suggest that the reported network

alterations in overexpressing models (Busche & Konnerth, 2015; Palop et al., 2007; Palop &

Mucke, 2010) might be due to overexpression of APP, rather than Aβ. Furthermore, we consider

that any results obtained with any overexpressing model on network dysfunction should be

evaluated against the new insights obtained by using the App KI models. It has been suggested that

more than 3000 publications using APP and APP/PS overexpressing models should be re-

evaluated (Saito et al., 2014a; Saito, Matsuba, Yamazaki, Hashimoto, & Saido, 2016). We believe

General Discussion

P a g e 92 | 128

the use of relevant controls, such as rodents overexpressing WT human proteins with WT protein

levels at the same level as the mutant protein should be implemented when using APP and APP/PS

overexpressing models. Furthermore, models such as the App KI mice are a good alternative to

investigate the effects of Aβ without overexpression artifacts. Although the App KI mice have been

extensively used since they became available, the combination of several familial mutations

remains artificial, and it could cause effects that are not present in sAD. For any model, it is

important to remind ourselves of its limitations to correctly interpret results.

Another objective of this thesis was to assess PAC as a sensitive functional readout of Aβ

pathology in the preclinical phase of AD in the two mentioned models. We focused our analysis

on PAC, as this measurement has been suggested to be associated with normal cognitive

functioning in humans (Axmacher et al., 2010; Canolty et al., 2006; Lega et al., 2016) and rodents

(Amemiya & Redish, 2018; Belluscio, Mizuseki, Schmidt, Kempter, & Buzsáki, 2012; Tort et al.,

2009, 2008), and to be disrupted in patients with AD and MCI (Dimitriadis et al., 2015; Poza et

al., 2017). Furthermore, it has been shown that injections of tau aggregated in the hippocampus

causes early PAC deficits in the P301L tau mouse model of AD pathology (Ahnaou et al., 2017).

Contrary to what we expected, our results indicate no changes in PAC at this early stage of

pathology. Although the studies carried out by Poza et al., and Dimitriadis et al., provide valuable

insights to better understand coupling changes associated with the disease in patients, both studies

reported PAC alterations after the preclinical phase. Based on previous literature (Tort et al., 2009,

2008), we expected to see PAC changes in a task-related manner. Opposite to this, for the TUNL

task, we did not observe differential theta-gamma coupling between the sample and choice stages

on the studied brain areas. A plausible explanation is that the selected brain areas are equally

involved in the two aspects of the task (sample and choice); therefore, we did not detect a

difference. Another explanation is related to the high accuracy level observed during the TUNL

task. For instance, Tort et al., (2009), demonstrated an increase in theta-gamma PAC in the

hippocampus as the rats learned an item-context association task, but once the performance was

high and stable, the strength in the coupling was maintained. Similarly, we only observed subtle

changes in PAC for WT mice during the VD task. Overall, it might be that the selected tasks in

this thesis are not sensitive to detect PAC changes. Research of other functional connectivity

analyzes investigated in patients (Engels et al., 2017) were outside the scope of this thesis, but we

believe they should be further explored in the different animal models used in the field.

General Discussion

P a g e 93 | 128

Altogether, our results do not support the hypothesis of early alterations in oscillations due

to Aβ amyloidosis. Future studies should focus on the investigation of electrophysiological

readouts using relevant models, like the App KI model. We believed the App KI and the MAPT KI

models could help to gain insights and outline a timeline on network changes associated with AD

pathology. In AD patients changes in power spectral or synchronization have been commonly

reported (Coben et al., 1985; Engels et al., 2015; Nimmrich et al., 2015; Voevodskaya et al., 2018).

Although some of these studies used new guidelines for diagnosis and investigated preclinical and

clinical AD, the reliability of the AD diagnosis and early detection in other studies must be

considered with caution, given the lack of available biomarkers at the time of the studies.

Electrophysiological readouts provide high temporal measurements of neuronal network

functioning that can be implemented in experimental models, but also in patients (Babiloni et al.,

2020). A key aspect to evaluate the potential value of electrophysiological readouts as biomarkers

is to understand when neuronal network alterations appear. If these alterations are present early on

during the preclinical AD phase, these readouts could have a major impact on disease diagnostics.

P a g e 94 | 128

References

P a g e 95 | 128

References

Abraha, A., Ghoshal, N., Gamblin, T. C., Cryns, V., Berry, R. W., Kuret, J., & Binder, L. I.

(2000). C-terminal inhibition of tau assembly in vitro and in Alzheimer’s disease. Journal

of Cell Science, 113(21), 3737–3745.

Ahnaou, A., Moechars, D., Raeymaekers, L., Biermans, R., Manyakov, N. V, Bottelbergs, A., …

Drinkenburg, W. H. (2017). Emergence of early alterations in network oscillations and

functional connectivity in a tau seeding mouse model of Alzheimer’s disease pathology.

Scientific Reports, 7(1), 1–14. https://doi.org/10.1038/s41598-017-13839-6

Alekseichuk, I., Turi, Z., Amador de Lara, G., Antal, A., & Paulus, W. (2016). Spatial Working

Memory in Humans Depends on Theta and High Gamma Synchronization in the Prefrontal

Cortex. Current Biology, 26(12), 1513–1521. https://doi.org/10.1016/j.cub.2016.04.035

Alzheimer’s Association, 2019. (2019). Alzheimer’s Disease Facts and Figures. Alzheimers

Dement 2019; 15 (3):321-87, 87. Retrieved from

https://www.alz.org/media/Documents/alzheimers-facts-and-figures-2019-r.pdf

Amemiya, S., & Redish, A. D. (2018). Hippocampal Theta-Gamma Coupling Reflects State-

Dependent Information Processing in Decision Making. Cell Reports, 22(12), 3328–3338.

https://doi.org/10.1016/j.celrep.2018.02.091

Apostolova. (2013). Hippocampal atrophy and ventricular enlargement in normal aging, 26(1),

17–27. https://doi.org/10.1097/WAD.0b013e3182163b62.Hippocampal

Axmacher, N., Henseler, M. M., Jensen, O., Weinreich, I., Elger, C. E., & Fell, J. (2010). Cross-

frequency coupling supports multi-item working memory in the human hippocampus.

Proceedings of the National Academy of Sciences, 107(7), 3228–3233.

https://doi.org/10.1073/pnas.0911531107

Babiloni, C., Blinowska, K., Bonanni, L., Cichocki, A., De Haan, W., Del Percio, C., … Randall,

F. (2020). What electrophysiology tells us about Alzheimer’s disease: a window into the

synchronization and connectivity of brain neurons. Neurobiology of Aging, 85, 58–73.

https://doi.org/10.1016/j.neurobiolaging.2019.09.008

References

P a g e 96 | 128

Başar, E., Schmiedt-Fehr, C., Mathes, B., Femir, B., Emek-Savaş, D. D., Tülay, E., … Başar-

Eroğlu, C. (2016). What does the broken brain say to the neuroscientist? Oscillations and

connectivity in schizophrenia, Alzheimer’s disease, and bipolar disorder. International

Journal of Psychophysiology, 103, 135–148. https://doi.org/10.1016/j.ijpsycho.2015.02.004

Bazzigaluppi, P., Beckett, T. L., Koletar, M. M., Lai, A. Y., Joo, I. L., Brown, M. E., …

Stefanovic, B. (2018). Early‐stage attenuation of phase‐amplitude coupling in the

hippocampus and medial prefrontal cortex in a transgenic rat model of Alzheimer’s disease.

Journal of Neurochemistry, 144(5), 669–679.

Bekris, L. M., Yu, C. E., Bird, T. D., & Tsuang, D. W. (2010, December). Review article:

Genetics of Alzheimer disease. Journal of Geriatric Psychiatry and Neurology. SAGE

PublicationsSage CA: Los Angeles, CA. https://doi.org/10.1177/0891988710383571

Belluscio, M. A., Mizuseki, K., Schmidt, R., Kempter, R., & Buzsáki, G. (2012). Cross-

frequency phase-phase coupling between θ and γ oscillations in the hippocampus. The

Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 32(2), 423–

435. https://doi.org/10.1523/JNEUROSCI.4122-11.2012

Benchenane, K., Tiesinga, P. H., & Battaglia, F. P. (2011). Oscillations in the prefrontal cortex:

A gateway to memory and attention. Current Opinion in Neurobiology. Elsevier Ltd.

https://doi.org/10.1016/j.conb.2011.01.004

Benjamini, Y., & Hochberg, Y. (1995). Controlling the False Discovery Rate: A Practical and

Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. Series B

(Methodological), 289–300.

Bierer, L. M., Hof, P. R., Purohit, D. P., Carlin, L., Schmeidler, J., Davis, K. L., & Perl, D. P.

(1995). Neocortical Neurofibrillary Tangles Correlate With Dementia Severity in

Alzheimer’s Disease. Archives of Neurology, 52(1), 81–88.

https://doi.org/10.1001/archneur.1995.00540250089017

Born, H. A., Kim, J.-Y., Savjani, R. R., Das, P., Dabaghian, Y. A., Guo, Q., … Jankowsky, J. L.

(2014). Genetic Suppression of Transgenic APP Rescues Hypersynchronous Network

Activity in a Mouse Model of Alzeimer’s Disease. Journal of Neuroscience, 34(11), 3826–

References

P a g e 97 | 128

3840. https://doi.org/10.1523/jneurosci.5171-13.2014

Braak, H., & Braak, E. (1991). Neuropathological stageing of Alzheimer-related changes. Acta

Neuropathologica, 82(4), 239–259. https://doi.org/10.1007/BF00308809

Brown, R., Lam, A. D., Gonzalez-Sulser, A., Ying, A., Jones, M., Chou, R. C., … Oren, I.

(2018). Circadian and Brain State Modulation of Network Hyperexcitability in Alzheimer’s

Disease. Eneuro, 5(2), ENEURO.0426-17.2018. https://doi.org/10.1523/ENEURO.0426-

17.2018

Busche, M. A., & Konnerth, A. (2015). Neuronal hyperactivity - A key defect in Alzeimer’s

disease? Bioessays Journal, 37, 624–632. https://doi.org/10.1002/bies.201500004

Buzsáki, G., & Draguhn, A. (2004). Neuronal oscillations in cortical networks. Science (New

York, N.Y.), 304(5679), 1926–1929. https://doi.org/10.1126/science.1099745

Canolty, R. T., Edwards, E., Dalal, S. S., Soltani, M., Nagarajan, S. S., Kirsch, H. E., … Knight,

R. T. (2006). High Gamma Power Is Phase-Locked to Theta Oscillations in Human

Neocortex. Science, 313(5793), 1626–1628. https://doi.org/10.1126/science.1128115

Citron, M. (2004). Strategies for disease modification in Alzheimer’s disease. Nature Reviews

Neuroscience, 5(9), 677–685. https://doi.org/10.1038/nrn1495

Coben, L. A., Danziger, W., & Storandt, M. (1985). A longitudinal EEG study of mild senile

dementia of Alzheimer type: changes at 1 year and at 2.5 years. Electroencephalography

and Clinical Neurophysiology, 61(2), 101–112. https://doi.org/10.1016/0013-

4694(85)91048-X

Cohen, M. X. (2014). A neural microcircuit for cognitive conflict detection and signaling. Trends

in Neurosciences, 37(9), 480–490. https://doi.org/10.1016/j.tins.2014.06.004

Cohen, R., Rezai-Zadeh, K., Weitz, T. M., Rentsendorj, A., Gate, D., Spivak, I., … Town, T.

(2013). A Transgenic Alzheimer Rat with Plaques, Tau Pathology, Behavioral Impairment,

Oligomeric A , and Frank Neuronal Loss. Journal of Neuroscience, 33(15), 6245–6256.

https://doi.org/10.1523/JNEUROSCI.3672-12.2013

Colgin, L. L., Denninger, T., Fyhn, M., Hafting, T., Bonnevie, T., Jensen, O., … Moser, E. I.

References

P a g e 98 | 128

(2009). Frequency of gamma oscillations routes flow of information in the hippocampus.

Nature, 462, 353. Retrieved from https://doi.org/10.1038/nature08573

Cornelissen, F., Cik, M., & Gustin, E. (2012). Phaedra, a Protocol-Driven System for Analysis

and Validation of High-Content Imaging and Flow Cytometry. Journal of Biomolecular

Screening, 17(4), 496–506. https://doi.org/10.1177/1087057111432885

De Strooper, B., & Karran, E. (2016). The Cellular Phase of Alzheimer’s Disease. Cell, 164(4),

603–615. https://doi.org/10.1016/j.cell.2015.12.056

DeKosky, S. T., & Scheff, S. W. (1990). Synapse loss in frontal cortex biopsies in Alzheimer’s

disease: Correlation with cognitive severity. Annals of Neurology, 27(5), 457–464.

https://doi.org/10.1002/ana.410270502

Dimitriadis, S. I., Laskaris, N. A., Bitzidou, M. P., Tarnanas, I., & Tsolaki, M. N. (2015). A

novel biomarker of amnestic MCI based on dynamic cross-frequency coupling patterns

during cognitive brain responses. Frontiers in Neuroscience, 9(OCT), 1–17.

https://doi.org/10.3389/fnins.2015.00350

Do Carmo, S., & Cuello, A. (2013). Modeling Alzheimer’s disease in transgenic rats. Molecular

Neurodegeneration, 8(1), 37. https://doi.org/10.1186/1750-1326-8-37

Drummond, E., & Wisniewski, T. (2017). Alzheimer’s disease: experimental models and reality.

Acta Neuropathologica. https://doi.org/10.1007/s00401-016-1662-x

Dubois, B., Hampel, H., Feldman, H. H., Scheltens, P., Aisen, P., Andrieu, S., … Jack, C. R.

(2016). Preclinical Alzheimer’s disease: Definition, natural history, and diagnostic criteria.

Alzheimer’s & Dementia, 12(3), 292–323. https://doi.org/10.1016/j.jalz.2016.02.002

Ellenbroek, B., & Youn, J. (2016). Rodent models in neuroscience research: is it a rat race?

Disease Models & Mechanisms, 9(10), 1079–1087. https://doi.org/10.1242/dmm.026120

Engels, M. M. A., Stam, C. J., Van der Flier, W. M., Scheltens, P., De Waal, H., & Van Straaten,

E. C. (2015). Declining functional connectivity and changing hub locations in Alzheimer’s

disease: an EEG study. BMC Neurology, 15(1), 145. https://doi.org/10.1186/s12883-015-

0400-7

References

P a g e 99 | 128

Engels, M. M. A., van der Flier, W. M., Stam, C. J., Hillebrand, A., Scheltens, P., & van

Straaten, E. C. W. (2017). Alzheimer’s disease: The state of the art in resting-state

magnetoencephalography. Clinical Neurophysiology, 128(8), 1426–1437.

https://doi.org/10.1016/j.clinph.2017.05.012

Fischer, C. E., Qian, W., Schweizer, T. A., Ismail, Z., Smith, E. E., Millikin, C. P., & Munoz, D.

G. (2017). Determining the impact of psychosis on rates of false-positive and false-negative

diagnosis in Alzheimer’s disease. In Alzheimer’s and Dementia: Translational Research

and Clinical Interventions (Vol. 3, pp. 385–392). Elsevier Inc.

https://doi.org/10.1016/j.trci.2017.06.001

Franklin, K. B., & Paxinos, G. (1997). The mouse brain in stereotaxic coordinates. Waltham:

Academic press.

Friedman, D., Honig, L. S., & Scarmeas, N. (2013). Seizures and Epilepsy in Alzheimer’s

Disease. CNS Neurosci Ther., 18(4), 285–294. https://doi.org/10.1111/j.1755-

5949.2011.00251.x.Seizures

Fries, P. (2015). Rhythms for Cognition: Communication through Coherence. Neuron, 88(1),

220–235. https://doi.org/10.1016/j.neuron.2015.09.034

Galeano, P., Martino Adami, P. V., Do Carmo, S., Blanco, E., Rotondaro, C., Capani, F., …

Morelli, L. (2014). Longitudinal analysis of the behavioral phenotype in a novel transgenic

rat model of early stages of Alzheimer’s disease. Frontiers in Behavioral Neuroscience,

8(September), 1–15. https://doi.org/10.3389/fnbeh.2014.00321

Games, D., Adams, D., Alessandrini, R., Barbour, R., Borthelette, P., Blackwell, C., … Zhao, J.

(1995). Alzheimer-type neuropathology in transgenic mice overexpressing V717F β-

amyloid precursor protein. Nature, 373(6514), 523–527. https://doi.org/10.1038/373523a0

Glenner, G. G., & Wong, C. W. (1984). Alzheimer’s disease: Initial report of the purification and

characterization of a novel cerebrovascular amyloid protein. Biochemical and Biophysical

Research Communications, 120(3), 885–890. https://doi.org/10.1016/S0006-

291X(84)80190-4

Glickman, M. E., Rao, S. R., & Schultz, M. R. (2014). False discovery rate control is a

References

P a g e 100 | 128

recommended alternative to Bonferroni-type adjustments in health studies. Journal of

Clinical Epidemiology. Elsevier Inc. https://doi.org/10.1016/j.jclinepi.2014.03.012

Goate, A., Chartier-Harlin, M.-C., Mullan, M., Brown, J., Crawford, F., Fidani, L., … Hardy, J.

(1991). Segregation of a missense mutation in the amyloid precursor protein gene with

familial Alzheimer’s disease. Nature, 349(6311), 704–706.

https://doi.org/10.1038/349704a0

Goutagny, R., Gu, N., Cavanagh, C., Jackson, J., & Chabot, J. (2013). Alterations in

hippocampal network oscillations and theta – gamma coupling arise before A b

overproduction in a mouse model of Alzheimer ’ s disease, 37(March), 1896–1902.

https://doi.org/10.1111/ejn.12233

Grant, S. M., Ducatenzeiler, A., Szyf, M., & Cuello, A. C. (2000). Aß Immunoreactive Material

Is Present in Several Intracellular Compartments in Transfected, Neuronally Differentiated,

P19 Cells Expressing the Human Amyloid ß-Protein Precursor. Journal of Alzheimer’s

Disease, 2(3–4), 207–222. https://doi.org/10.3233/JAD-2000-23-403

Grundke-Iqbal, I., Iqbal, K., Tung, Y. C., Quinlan, M., Wisniewski, H. M., & Binder, L. I.

(1986). Abnormal phosphorylation of the microtubule-associated protein tau (tau) in

Alzheimer cytoskeletal pathology. Proceedings of the National Academy of Sciences,

83(13), 4913–4917. https://doi.org/10.1073/pnas.83.13.4913

Hall, A. M., & Roberson, E. D. (2012). Mouse models of Alzheimer’s disease. Brain Research

Bulletin, 88(1), 3–12. https://doi.org/10.1016/j.brainresbull.2011.11.017

Hardy, J., & Higgins, G. (1992). Alzheimer’s disease: the amyloid cascade hypothesis. Science,

256(5054), 184–185. https://doi.org/10.1126/science.1566067

Hasselmo, M. E., & Stern, C. E. (2014). Theta rhythm and the encoding and retrieval of space

and time. NeuroImage, 85, 656–666. https://doi.org/10.1016/j.neuroimage.2013.06.022

Herrmann, C. S., Strüber, D., Helfrich, R. F., & Engel, A. K. (2016). EEG oscillations: From

correlation to causality. International Journal of Psychophysiology, 103, 12–21.

https://doi.org/10.1016/j.ijpsycho.2015.02.003

Höfling, C., Morawski, M., Zeitschel, U., Zanier, E. R., Moschke, K., Serdaroglu, A., … Kuhn,

References

P a g e 101 | 128

P.-H. (2016). Differential transgene expression patterns in Alzheimer mouse models

revealed by novel human amyloid precursor protein-specific antibodies. Aging Cell, 15(5),

953–963. https://doi.org/10.1111/acel.12508

Holcomb, L., Gordon, M. N., McGowan, E., Yu, X., Benkovic, S., Jantzen, P., … Duff, K.

(1998). Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant

amyloid precursor protein and presenilin 1 transgenes. Nature Medicine, 4(1), 97–100.

https://doi.org/10.1038/nm0198-097

Horner, A. E., Heath, C. J., Hvoslef-Eide, M., Kent, B. A., Kim, C. H., Nilsson, S. R. O., …

Bussey, T. J. (2013). The touchscreen operant platform for testing learning and memory in

rats and mice. Nature Protocols, 8(10), 1961–1984. https://doi.org/10.1038/nprot.2013.122

Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y., Younkin, S., … Cole, G. (1996).

Correlative Memory Deficits, A Elevation, and Amyloid Plaques in Transgenic Mice.

Science, 274(5284), 99–103. https://doi.org/10.1126/science.274.5284.99

Ittner, L. M., & Götz, J. (2011). Amyloid-β and tau — a toxic pas de deux in Alzheimer’s

disease. Nature Reviews Neuroscience, 12(2), 67–72. https://doi.org/10.1038/nrn2967

Iulita, M. F., Allard, S., Richter, L., Munter, L.-M., Ducatenzeiler, A., Weise, C., … Cuello, A.

C. (2014). Intracellular Aβ pathology and early cognitive impairments in a transgenic rat

overexpressing human amyloid precursor protein: a multidimensional study. Acta

Neuropathologica Communications, 2(1), 61. https://doi.org/10.1186/2051-5960-2-61

Iulita, M. F., Bistué Millón, M. B., Pentz, R., Aguilar, L. F., Do Carmo, S., Allard, S., … Cuello,

A. C. (2017). Differential deregulation of NGF and BDNF neurotrophins in a transgenic rat

model of Alzheimer’s disease. Neurobiology of Disease, 108(February), 307–323.

https://doi.org/10.1016/j.nbd.2017.08.019

Jack, C. R., Hampel, H. J., Universities, S., Cu, M., & Petersen, R. C. (2016). A new

classification system for AD , independent of cognition A / T / N : An unbiased descriptive

classification scheme for Alzheimer disease biomarkers. Neurology, 0(July), 1–10.

Jack, C. R., Knopman, D. S., Jagust, W. J., Shaw, L. M., Aisen, P. S., Weiner, M. W., …

Trojanowski, J. Q. (2010). Hypothetical model of dynamic biomarkers of the Alzheimer’s

References

P a g e 102 | 128

pathological cascade. The Lancet Neurology, 9(1), 119–128. https://doi.org/10.1016/S1474-

4422(09)70299-6

Jankowsky, J. L., & Zheng, H. (2017). Practical considerations for choosing a mouse model of

Alzheimer’s disease. Molecular Neurodegeneration, 1–22. https://doi.org/10.1186/s13024-

017-0231-7

Jobert, M., Wilson, F. J., Ruigt, G. S. F., Brunovsky, M., Prichep, L. S., & Drinkenburg, W. H. I.

M. (2012). Guidelines for the Recording and Evaluation of Pharmaco-EEG Data in Man:

The International Pharmaco-EEG Society (IPEG). Neuropsychobiology, 66(4), 201–220.

https://doi.org/10.1159/000343478

Kalimo, H., Lalowski, M., Bogdanovic, N., Philipson, O., Bird, T. D., Nochlin, D., … Ingelsson,

M. (2013). The Arctic AβPP mutation leads to Alzheimer’s disease pathology with highly

variable topographic deposition of differentially truncated Aβ. Acta Neuropathologica

Communications, 1(1), 60. https://doi.org/10.1186/2051-5960-1-60

Karch, C. M., & Goate, A. M. (2015). Alzheimer’s Disease Risk Genes and Mechanisms of

Disease Pathogenesis. Biological Psychiatry, 77(1), 43–51.

https://doi.org/10.1016/j.biopsych.2014.05.006

Karran, E., & De Strooper, B. (2016). The amyloid cascade hypothesis: are we poised for

success or failure? Journal of Neurochemistry, 139, 237–252.

https://doi.org/10.1111/jnc.13632

Karran, E., Mercken, M., & De Strooper, B. (2011). The amyloid cascade hypothesis for

Alzheimer’s disease: an appraisal for the development of therapeutics. Nature Reviews

Drug Discovery, 10(9), 698–712. https://doi.org/10.1038/nrd3505

Keren, G. (2014). Between-or within-subjects design: A methodological dilemma. A Handbook

for Data Analysis in the Behaviorial Sciences.

Kiyota, T. (2014). Animal Models of Alzheimer’s Disease. In H. Xiong & H. E. Gendelman

(Eds.) (pp. 527–540). New York, NY: Springer New York. https://doi.org/10.1007/978-1-

4614-8794-4_35

LaFerla, F. M., & Green, K. N. (2012). Animal models of Alzheimer disease. Cold Spring Harb

References

P a g e 103 | 128

Perspect Med, 2(11), 1–13. https://doi.org/10.1101/cshperspect.a006320

Latif-Hernandez, A., Shah, D., Craessaerts, K., Saido, T., Saito, T., De Strooper, B., …

D’Hooge, R. (2019). Subtle behavioral changes and increased prefrontal-hippocampal

network synchronicity in APP NL−G−F mice before prominent plaque deposition.

Behavioural Brain Research, 364(November 2017), 431–441.

https://doi.org/10.1016/j.bbr.2017.11.017

Lega, B., Burke, J., Jacobs, J., & Kahana, M. J. (2016). Slow-Theta-to-Gamma Phase-Amplitude

Coupling in Human Hippocampus Supports the Formation of New Episodic Memories.

Cerebral Cortex, 26(1), 268–278. https://doi.org/10.1093/cercor/bhu232

Leon, W. C., Canneva, F., Partridge, V., Allard, S., Ferretti, M. T., Dewilde, A., … Cuello, A. C.

(2010). A novel transgenic rat model with a full alzheimer’s - Like amyloid pathology

displays pre - Plaque intracellular amyloid -beta- Associated cognitive impairment. Journal

of Alzheimer’s Disease, 20(1), 113–126. https://doi.org/10.3233/JAD-2010-1349

Lindwall, G., & Cole, R. D. (1984). Phosphorylation affects the ability of tau protein to promote

microtubule assembly. Journal of Biological Chemistry, 259(8), 5301–5305.

Lisman, J. E., & Jensen, O. (2013). The Theta-Gamma Neural Code. Neuron. Elsevier Inc.

https://doi.org/10.1016/j.neuron.2013.03.007

Livingston, G., Sommerlad, A., Orgeta, V., Costafreda, S. G., Huntley, J., Ames, D., …

Mukadam, N. (2017). Dementia prevention, intervention, and care. Lancet (London,

England), 390(10113), 2673–2734. https://doi.org/10.1016/S0140-6736(17)31363-6

Lott, I. T. (2012). Neurological phenotypes for Down syndrome across the life span (pp. 101–

121). https://doi.org/10.1016/B978-0-444-54299-1.00006-6

Mandelkow, E., Stamer, K., Vogel, E., & Thies, E. (2003). Clogging of axons by tau, inhibition

of axonal traffic and starvation of synapses. Neurobiology of Aging, 24(8), 1079–1085.


Masuda, A., Kobayashi, Y., Kogo, N., Saito, T., Saido, T. C., & Itohara, S. (2016). Cognitive

deficits in single App knock-in mouse models. Neurobiology of Learning and Memory, 135,

73–82. https://doi.org/10.1016/j.nlm.2016.07.001

References

P a g e 104 | 128

Maya, N., & Bassem, H. A. (2014). Amyloid precursor protein and neural development.

Development, 141(13), 2543–2548. https://doi.org/10.1242/dev.108712

Mehla, J., Lacoursiere, S. G., Lapointe, V., Mcnaughton, B. L., Sutherland, R. J., Mcdonald, R.

J., & Mohajerani, M. H. (2019). Age-dependent behavioral and biochemical

characterization of single APP knock-in mouse ( APP NL-G-F / NL-G-F ) model of

Alzheimer ’ s disease. Neurobiology of Aging, 75, 25–37.


Mockett, B. G., Richter, M., Abraham, W. C., & Müller, U. C. (2017). Therapeutic Potential of

Secreted Amyloid Precursor Protein APPsα. Frontiers in Molecular Neuroscience, 10.

https://doi.org/10.3389/fnmol.2017.00030

Mucke, L., Masliah, E., Yu, G.-Q., Mallory, M., Rockenstein, E. M., Tatsuno, G., …

McConlogue, L. (2000). High-Level Neuronal Expression of Aβ 1–42 in Wild-Type Human

Amyloid Protein Precursor Transgenic Mice: Synaptotoxicity without Plaque Formation.

The Journal of Neuroscience, 20(11), 4050–4058. https://doi.org/10.1523/JNEUROSCI.20-

11-04050.2000

Murrell, J., Farlow, M., Ghetti, B., & Benson, M. (1991). A mutation in the amyloid precursor

protein associated with hereditary Alzheimer’s disease. Science, 254(5028), 97–99.

https://doi.org/10.1126/science.1925564

Nilsson, P., Saito, T., & Saido, T. C. (2014). New Mouse Model of Alzheimer’s. ACS Chemical

Neuroscience, 5(7), 499–502. https://doi.org/10.1021/cn500105p

Nimmrich, V., Draguhn, A., & Axmacher, N. (2015). Neuronal Network Oscillations in

Neurodegenerative Diseases. NeuroMolecular Medicine, 17(3), 270–284.

https://doi.org/10.1007/s12017-015-8355-9

Oakley, H., Cole, S. L., Logan, S., Maus, E., Shao, P., Craft, J., … Vassar, R. (2006).

Intraneuronal beta-Amyloid Aggregates, Neurodegeneration, and Neuron Loss in

Transgenic Mice with Five Familial Alzheimer’s Disease Mutations: Potential Factors in

Amyloid Plaque Formation. Journal of Neuroscience, 26(40), 10129–10140.

https://doi.org/10.1523/JNEUROSCI.1202-06.2006

References

P a g e 105 | 128

Oomen, C. A., Hvoslef-Eide, M., Heath, C. J., Mar, A. C., Horner, A. E., Bussey, T. J., &

Saksida, L. M. (2013). The touchscreen operant platform for testing working memory and

pattern separation in rats and mice. Nature Protocols, 8(10), 2006–2021.

https://doi.org/10.1038/nprot.2013.124

Palop, J. J., Chin, J., Roberson, E. D., Wang, J., Thwin, M. T., Bien-Ly, N., … Mucke, L.

(2007). Aberrant Excitatory Neuronal Activity and Compensatory Remodeling of Inhibitory

Hippocampal Circuits in Mouse Models of Alzheimer’s Disease. Neuron, 55(5), 697–711.

https://doi.org/10.1016/j.neuron.2007.07.025

Palop, J. J., & Mucke, L. (2009). Epilepsy and Cognitive Impairments in Alzheimer Disease.

Arch Neurol., 66(4). https://doi.org/10.1001/archeneurol.2009.15.

Palop, J. J., & Mucke, L. (2010). Amyloid-Β-induced neuronal dysfunction in Alzheimer’s

disease: From synapses toward neural networks. Nature Neuroscience, 13(7), 812–818.

https://doi.org/10.1038/nn.2583

Palop, J. J., & Mucke, L. (2016). Network abnormalities and interneuron dysfunction in

Alzheimer disease. Nature Reviews Neuroscience, 17(12), 777–792.

https://doi.org/10.1038/nrn.2016.141

Parent, M. J., Zimmer, E. R., Shin, M., Kang, M. S., Fonov, V. S., Mathieu, A., … Rosa-Neto, P.

(2017). Multimodal imaging in rat model recapitulates Alzheimer’s Disease biomarkers

abnormalities. The Journal of Neuroscience : The Official Journal of the Society for

Neuroscience. https://doi.org/10.1523/JNEUROSCI.1346-17.2017

Parihar, M. S., & Brewer, G. J. (2010). Amyloid-β as a Modulator of Synaptic Plasticity. Journal

of Alzheimer’s Disease, 22(3), 741–763. https://doi.org/10.3233/JAD-2010-101020

Paxinos, G., & Watson, C. (1998). The Rat Brain in Stereotaxic Coordinates (4th editio).

Pontecorvo, M. J., Devous, M. D., Navitsky, M., Lu, M., Salloway, S., Schaerf, F. W., …

Mintun, M. A. (2017). Relationships between flortaucipir PET tau binding and amyloid

burden, clinical diagnosis, age and cognition. Brain, aww334.

https://doi.org/10.1093/brain/aww334

Poza, J., Bachiller, A., Gómez, C., García, M., Nunez, P., Gomez-Pilar, J., … Roberto, H.

References

P a g e 106 | 128

(2017). Phase-Amplitude Coupling Analysis of Spontaneous EEG Activity in Alzheimer’s

Disease. IEEE, 2259–2262. https://doi.org/10.1109/EMBC.2017.8037305

Qi, Y., Klyubin, I., Harney, S. C., Hu, N., Cullen, W. K., Grant, M. K., … Rowan, M. J. (2014).

Longitudinal testing of hippocampal plasticity reveals the onset and maintenance of

endogenous human Aß-induced synaptic dysfunction in individual freely behaving pre-

plaque transgenic rats: rapid reversal by anti-Aß agents. Acta Neuropathologica

Communications, 2(1), 175. https://doi.org/10.1186/s40478-014-0175-x

Querfurth, H. M., & LaFerla, F. M. (2011). Mechanisms of Disease Alzheimer’s. The New Engl

and Journal of Medicine, 56(3), 687–696. https://doi.org/10.1056/NEJMra0909142

Rice, H. C., De Malmazet, D., Schreurs, A., Frere, S., Van Molle, I., Volkov, A. N., … De Wit,

J. (2019). Secreted amyloid-b precursor protein functions as a GABA B R1a ligand to

modulate synaptic transmission. Science, 363(6423).

https://doi.org/10.1126/science.aao4827

Saito, T., Matsuba, Y., Mihira, N., Takano, J., Nilsson, P., Itohara, S., … Saido, T. C. (2014a).

Single App knock-in mouse models of Alzheimer’s disease. Nature Neuroscience, 17(5),

661–663. https://doi.org/10.1038/nn.3697

Saito, T., Matsuba, Y., Mihira, N., Takano, J., Nilsson, P., Itohara, S., … Saido, T. C. (2014b).

Single App knock-in mouse models of Alzheimer’s disease. Nature Neuroscience, 17(5),

661–663. https://doi.org/10.1038/nn.3697

Saito, T., Matsuba, Y., Yamazaki, N., Hashimoto, S., & Saido, T. C. (2016). Calpain Activation

in Alzheimer’s Model Mice Is an Artifact of APP and Presenilin Overexpression. Journal of

Neuroscience, 36(38), 9933–9936. https://doi.org/10.1523/jneurosci.1907-16.2016

Saito, T., Mihira, N., Matsuba, Y., Sasaguri, H., Hashimoto, S., Narasimhan, S., … Saido, T. C.

(2019). Humanization of the entire murine Mapt gene provides a murine model of

pathological human tau propagation . Journal of Biological Chemistry, 294(34), 12754–

12765. https://doi.org/10.1074/jbc.ra119.009487

Sakakibara, Y., Sekiya, M., Saito, T., Saido, T. C., & Iijima, K. M. (2018). Cognitive and

emotional alterations in App knock-in mouse models of Aβ amyloidosis. BMC

References

P a g e 107 | 128

Neuroscience, 19(1), 1–17. https://doi.org/10.1186/s12868-018-0446-8

Sasaguri, H., Nilsson, P., Hashimoto, S., Nagata, K., Saito, T., De Strooper, B., … Saido, T. C.

(2017). APP mouse models for Alzheimer’s disease preclinical studies. The EMBO Journal,

36(17), e201797397. https://doi.org/10.15252/embj.201797397

Scheff, S. W., Price, D. A., Schmitt, F. A., & Mufson, E. J. (2006). Hippocampal synaptic loss in

early Alzheimer’s disease and mild cognitive impairment. Neurobiology of Aging, 27(10),

1372–1384. https://doi.org/10.1016/j.neurobiolaging.2005.09.012

Scheltens, P., Blennow, K., Breteler, M. M. B., De Strooper, B., Frisoni, G. B., Salloway, S., &

Van der Flier, W. M. (2016). Alzheimer’s disease. The Lancet, 388(10043), 505–517.

https://doi.org/10.1016/S0140-6736(15)01124-1

Selkoe, D. J. (2002). Alzheimer’s Disease Is a Synaptic Failure. Science, 298(5594), 789–791.

https://doi.org/10.1126/science.1074069

Serrano-Pozo, A., Frosch, M. P., Masliah, E., & Hyman, B. T. (2011). Neuropathological

alterations in Alzheimer disease. Cold Spring Harbor Perspectives in Medicine, 1(1), 1–23.

https://doi.org/10.1101/cshperspect.a006189

Shafiei, S. S., Guerrero-Muñoz, M. J., & Castillo-Carranza, D. L. (2017). Tau Oligomers:

Cytotoxicity, Propagation, and Mitochondrial Damage. Frontiers in Aging Neuroscience, 9.

https://doi.org/10.3389/fnagi.2017.00083

Sperling, R. A., Aisen, P. S., Beckett, L. A., Bennett, D. A., Craft, S., Fagan, A. M., … Phelps,

C. H. (2011). Toward defining the preclinical stages of Alzheimer’s disease:

Recommendations from the National Institute on Aging-Alzheimer’s Association

workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s and Dementia,

7(3), 280–292. https://doi.org/10.1016/j.jalz.2011.03.003

Spillantini, M. G., & Goedert, M. (2013). Tau pathology and neurodegeneration. The Lancet

Neurology, 12(6), 609–622. https://doi.org/10.1016/S1474-4422(13)70090-5

Spinney, L. (2014). Alzheimer’s disease: The forgetting gene. Nature, 510(7503), 26–28.

https://doi.org/10.1038/510026a

References

P a g e 108 | 128

Stelzmann, R. A., Norman Schnitzlein, H., & Reed Murtagh, F. (1995). An english translation of

alzheimer’s 1907 paper, “über eine eigenartige erkankung der hirnrinde.” Clinical Anatomy,

8(6), 429–431. https://doi.org/10.1002/ca.980080612

Sturchler-Pierrat, C., Abramowski, D., Duke, M., Wiederhold, K.-H., Mistl, C., Rothacher, S., …

Sommer, B. (1997). Two amyloid precursor protein transgenic mouse models with

Alzheimer disease-like pathology. Proceedings of the National Academy of Sciences,

94(24), 13287–13292. https://doi.org/10.1073/pnas.94.24.13287

Talpos, J. C., McTighe, S. M., Dias, R., Saksida, L. M., & Bussey, T. J. (2010). Trial-unique,

delayed nonmatching-to-location (TUNL): A novel, highly hippocampus-dependent

automated touchscreen test of location memory and pattern separation. Neurobiology of

Learning and Memory, 94(3), 341–352. https://doi.org/10.1016/j.nlm.2010.07.006

Tamura, M., Spellman, T. J., Rosen, A. M., Gogos, J. A., & Gordon, J. A. (2017). Hippocampal-

prefrontal theta-gamma coupling during performance of a spatial working memory task.

Nature Communications, 8(1), 2182. https://doi.org/10.1038/s41467-017-02108-9

TCW, J., & Goate, A. M. (2017). Genetics of β-Amyloid Precursor Protein in Alzheimer’s

Disease. Cold Spring Harbor Perspectives in Medicine, 7(6), a024539.

https://doi.org/10.1101/cshperspect.a024539

Terry, R. D., Masliah, E., Salmon, D. P., Butters, N., DeTeresa, R., Hill, R., … Katzman, R.

(1991). Physical basis of cognitive alterations in alzheimer’s disease: Synapse loss is the

major correlate of cognitive impairment. Annals of Neurology, 30(4), 572–580.

https://doi.org/10.1002/ana.410300410

Terwel, D., Muyllaert, D., Dewachter, I., Borghgraef, P., Croes, S., Devijver, H., & Van Leuven,

F. (2008). Amyloid Activates GSK-3β to Aggravate Neuronal Tauopathy in Bigenic Mice.

The American Journal of Pathology, 172(3), 786–798.

https://doi.org/10.2353/ajpath.2008.070904

Thorn, C. A., Atallah, H., Howe, M., & Graybiel, A. M. (2010). Differential dynamics of activity

changes in dorsolateral and dorsomedial striatal loops during learning. Neuron, 66(5), 781–

795. https://doi.org/10.1016/j.neuron.2010.04.036

References

P a g e 109 | 128

Tort, A. B. L., Komorowski, R. W., Manns, J. R., Kopell, N. J., & Eichenbaum, H. (2009).

Theta-gamma coupling increases during the learning of item-context associations.

Proceedings of the National Academy of Sciences, 106(49), 20942–20947.


Tort, A. B. L., Kramer, M. A., Thorn, C., Gibson, D. J., Kubota, Y., Graybiel, A. M., & Kopell,

N. J. (2008). Dynamic cross-frequency couplings of local field potential oscillations in rat

striatum and hippocampus during performance of a T-maze task. Proceedings of the

National Academy of Sciences, 105(51), 20517–20522.


Visser, P. J., Vos, S., van Rossum, I., & Scheltens, P. (2012). Comparison of International

Working Group criteria and National Institute on Aging–Alzheimer’s Association criteria

for Alzheimer’s disease. Alzheimer’s & Dementia, 8(6), 560–563.

https://doi.org/10.1016/j.jalz.2011.10.008

Voevodskaya, O., Pereira, J. B., Volpe, G., Lindberg, O., Stomrud, E., van Westen, D., …

Hansson, O. (2018). Altered structural network organization in cognitively normal

individuals with amyloid pathology. Neurobiology of Aging, 64, 15–24.


Walsh, C., Drinkenburg, W. H. I. M., & Ahnaou, A. (2017). Neurophysiological assessment of

neural network plasticity and connectivity: Progress towards early functional biomarkers for

disease interception therapies in Alzheimer’s disease. Neuroscience and Biobehavioral

Reviews, 73, 340–358. https://doi.org/10.1016/j.neubiorev.2016.12.020

Weissgerber, T. L., Milic, N. M., Winham, S. J., & Garovic, V. D. (2015). Beyond bar and line

graphs: time for a new data presentation paradigm. PLoS Biology, 13(4), e1002128.

https://doi.org/10.1371/journal.pbio.1002128

Whyte, L. S., Hemsley, K. M., Lau, A. A., Hassiotis, S., Saito, T., Saido, T. C., … Sargeant, T. J.

(2018). Reduction in open field activity in the absence of memory deficits in the

AppNL−G−F knock-in mouse model of Alzheimer’s disease. Behavioural Brain Research,

336(August 2017), 177–181. https://doi.org/10.1016/j.bbr.2017.09.006

References

P a g e 110 | 128

Wilson, E. N., Abela, A. R., Do Carmo, S., Allard, S., Marks, A. R., Welikovitch, L. A., …

Cuello, A. C. (2016). Intraneuronal Amyloid Beta Accumulation Disrupts Hippocampal

CRTC1-Dependent Gene Expression and Cognitive Function in a Rat Model of Alzheimer

Disease. Cerebral Cortex, bhv332. https://doi.org/10.1093/cercor/bhv332

Wippold, F. J., Cairns, N., Vo, K., Holtzman, D. M., & Morris, J. C. (2008). Neuropathology for

the neuroradiologist: Plaques and tangles. American Journal of Neuroradiology, 29(1), 18–

22. https://doi.org/10.3174/ajnr.A0781

Yoshiyama, Y., Lee, V. M.-Y., & Trojanowski, J. Q. (2001). Frontotemporal dementia and

tauopathy. Current Neurology and Neuroscience Reports, 1(5), 413–421.

https://doi.org/10.1007/s11910-001-0100-0

Scientific Contributions

P a g e 111 | 128

Scientific Contributions

Scientific Acknowledgements and Personal Contribution

• The candidate, Sofia Jacob, designed, organized and performed all in vivo experiments.

• The Matlab toolbox for the electrophysiological analysis was created by Nikolay Manyakov.

• Immunohistochemical staining of the McGill-R-Thy1-APP rats were carried out by Kristel Buyens

and Astrid Bottelbergs.

• McSA1 antibody specificity experiment was carried out by Marianne Borgers.

• Genotyping of McGill-R-Thy1-APP rats using ddPCR was designed and performed by Louis De

Muynck.

• Brain Aβ1-42 ELISA and immunohistochemical experiments for the AppNL-G-F mice were designed

and performed by Marijke De Bock, Bart Hermans, Cindy Wintmolders, Astrid Bottelbergs,

Marianne Borgers, Clara Theunis, and Bianca Van Broeck.

Conflict of Interest

No competing interests to declare.

P a g e 112 | 128

Curriculum Vitae

P a g e 113 | 128

Curriculum Vitae

Sofia Jacob

[email protected]

Current Position:

Present – September 2019 – Modis, Belgium

Project Manager Life Sciences

Education:

Present – 2015 – PhD Student, Janssen Pharmaceutica, KU Leuven, Belgium

PhD Thesis submitted: “Neurophysiological Assessment of Brain Network Activity

Involved in Cognitive Processing in Animal Models of Alzheimer’s Disease”.

Promoters: Prof. Dr. Detlef Balschun and Prof. Dr. Wilhelmus H.I.M. (Pim) Drinkenburg

2015 – 2013 – Master in Neuroscience. Erasmus Mundus, Université de Bordeaux, France

Double diploma from Université de Bordeaux and Charité Universitätsmedizin, Berlin,

with a semester abroad at Vrije Universiteit, Amsterdam.

Master thesis in industry at Janssen Pharmaceutica, Belgium

Final grade: A = magna cum laude

2013 – 2009 – Bachelor of Science, Honors in Psychology, Behavioral Neuroscience. Concordia

University, Montreal, Canada

Degree conferred with distinction (final grade: A = magna cum laude)

Publications:

• Jacob, S., Davies, G., De Bock, M. Hermans, B., Wintmolders, C., Bottelbergs, A.,

Borgers, M., Theunis, C., Van Broeck, B., Manyakov N. V., Balschun, B., & Drinkenburg

W.H.I.M. (2019). Neural oscillations during cognitive processes in an App knock-in mouse

mailto:[email protected]

Curriculum Vitae

P a g e 114 | 128

model of Alzheimer’s disease pathology. Sci Rep 9, 16363 doi:10.1038/s41598-019-

51928-w

• Gervais, N. J., Jacob, S., Brake, W.G., & Mumby, D. G., (2014). Modulatory effect of 17-

β estradiol on performance of ovariectomized rats on the Shock-Probe test. Physiology and

Behaviour. doi: 10.1016/j.physbeh.2014.04.030

• Gervais, N. J., Jacob, S., Brake, W.G., & Mumby, D. G., (2013). Systemic and intra-rhinal-

cortical 17-β estradiol administration modulate object-recognition memory in

ovariectomized female rats. Hormones and Behaviour, 64, 4 doi:

10.1016/j.yhbeh.2013.08.010

Talks:

• Jacob, S., Ahnaou A., Drinkenburg, WH., (2017, September). Neurophysiological

Assessment of Network Plasticity and Connectivity in a Tau Preclinical Mouse Model of

Alzheimer’s disease. International Conference on Basic and Clinical Multimodal Imaging.

Bern, Switzerland

• Jacob, S., (2016, October) Advanced EEG imaging of neuronal network interactions

during spatial working memory performance in rats: Paving the road for pharmacological

assessments. 19th International Pharmaco-EEG society meeting, Nijmegen, The

Netherlands.

Relevant Posters:

• Jacob, S., Davies. G., Manyakov. N., Drinkenburg, WH., (2018, October).

Characterization of neural oscillations during cognitive processes in an APP knock-in

mouse model of Alzheimer’s disease pathology. Poster submitted to the 1st annual meeting

of the Janssen Benelux Postdoc/ PhD student community, Beerse, Belgium.

• Tahon, K., Jackson, DA., Jacob, S., Drinkenburg, WH., (2016, November). Hippocampal

sharp-wave ripple characteristics during delay periods in the rodent automated search task.

Curriculum Vitae

P a g e 115 | 128

Poster submitted to the 46th annual meeting of the Society for Neuroscience, San Diego,

California, United States.

• Tahon, K., Jackson, DA., Jacob, S., Drinkenburg, WH., (2016, October).

Neurophysiological substrates of memory process. 19th International Pharmaco-EEG

society meeting, Nijmegen, The Netherlands.

• Jacob, S., Tahon, K., Steckler, T., Drinkenburg, P., (2016, March). Glutamatergic

manipulation in spatial search processes: advanced cognitive neurophysiological analysis

in rats. Poster submitted to the ECNP workshop for junior scientists in Europe, Nice,

France.

• Jacob, S., Tahon, K., Drinkenburg, P., (2015, July). Effects of MK-801 and Memantine on

learning and memory in rats: Advanced electroencephalography and behavioral analysis

suing an automated spatial search task. Poster submitted to the Neurasmus annual meeting,

Amsterdam, The Netherlands.

Relevant Conferences and Courses:

• 1st Annual Meeting of the Janssen Benelux Post-Doc/ PhD Student Community, Beerse,

Belgium (2018, October).

• Advanced course on Circuit Dynamic and Cognition. Neuroscience school of advanced

studies. Coordinator: Gyorgy Buzsaki (2018, June).

• Advanced course on Learning and Memory: Cellular and Molecular Mechanisms.

Neuroscience school of advanced studies. Coordinators: Susumu Tonegawa & Alcino J.

Silva (2017, June).

• 13h International Conference on Alzheimer’s & Parkinson’s Diseases, Vienna, Austria

(2017, March)

• 46th Annual Meeting of the Society for Neuroscience, San Diego, California, United States

(2016, November)

• 19th International Pharmaco-EEG society meeting, Nijmegen, The Netherlands (2016,

October)

Curriculum Vitae

P a g e 116 | 128

• European College of Neuropsychopharmacology workshop for junior scientists in Europe,

Nice, France (2016, March)

• FELASA B - Laboratory animal science (2013, July)

Scholarships:

• IWT Baekeland Mandate (2019 – 2015)

• Erasmus Mundus Student Scholarship – Category A (2015 – 2013)

• Certificate of Academic Excellence – Canadian Psychology Association (2013, June)

Acknowledgements

Acknowledgements

Here I am, writing the last words of my thesis. Looking back, I find myself full of emotions.

Without a doubt, I have grown scientifically and personally during this journey, and I owe this

growth to many people to whom I would like to express my appreciation.

First, I would like to voice my sincere gratitude to my supervisors Prof. Dr. Pim

Drinkenburg and Prof. Dr. Detlef Balschun for giving me the opportunity to conduct my Ph.D. at

the KU Leuven and Janssen Pharmaceutica. Pim, thank you for the unconditional support you have

offered me during these years. I think we both remember very well our first telephone conversation

where I told you I was not sure if electrophysiology was for me…if I only knew that 6 months

later, we were going to be writing a project proposal to start a Ph.D. in that topic. Detlef, thank

you for sharing your knowledge in the field of Alzheimer’s disease and providing me with

insightful directions. Both of you have given me a lot of freedom to grow as a scientist, allowing

me to design my experiments and making my own decisions. Of course, this came with the cost of

bumping my head against the wall multiple times, but in the end, I grew stronger.

Next to my supervisors, I would like to thank Dr. Nikolay (Kolya) V. Manyakov. Kolya,

you did not only help me with the development of the Matlab toolbox and its numerous

modifications, but your scientific input has a great impact on the way I think. I learned to appreciate

the “rawness” of your communication, challenging what I was doing. Thank you for all your

support!

Probably I would have never started this adventure if it was not for Dr. Koen Tahon. Koen,

thank you for believing in me and for teaching me so much in such a short period of time. You

have become a good friend. I truly missed you in the lab during the second half of my Ph.D.

I am also very grateful to my thesis committee members, Prof. Dr. Cees Van Leeuwen,

Prof. Dr. Andreas Van Leupoldt, I appreciate your input and suggestion during the many

interactions we had during my Ph.D., as well as for correcting this manuscript. Dr. Per Nilsson,

Prof. Dr. Jos Prickaerts, thank you for taking the time to travel to Belgium to be part of the jury of

my thesis and reviewing this manuscript.

Janssen colleagues, thank you for the great time! To all my in vivo colleagues: you made

the many hours working in the dark very bright. Thank you to the LAM team for helping us to

provide the best care for the animals. Willy, dank je wel voor de Nederlandse oefeningen and the

Acknowledgements

nice chats! Thank you to all members of Pim’s team, Abdel, Koen, Marjolein, Sean, Ria, Sofie,

Heidi, Christ, Leen, and Annemie, for all your support and encouragement. Heidi, thank you for

introducing me to the Janssen running team, for always being there for me (in the lab, but

especially at a personal level), and for being so real. I truly admire your endurance and strength

and I am very happy you are part of my life. Sean, my office mate, thank you for the great scientific

discussions! To the Neuroscience in vitro department, I would like to take this opportunity to thank

Bianca, Clara, Marianne, Astrid, Bart and Louis for helping and teaching me in vitro techniques.

During my project I believe I re-did my analysis 100 times, thank you, Tom, for providing advice

on statistical analysis. During my Ph.D. I have the chance to supervise three master students. I

enjoyed this experience tremendously. Julia, Natalie and Gethin, I hope you enjoyed working

together as much as I did. Thank you to my lunch group who provided a great distraction to the

long days in the lab. Many of you have become great friends!

I am proud to be one of the founding members of the Benelux Janssen Postdoc and Ph.D.

student community. Thank you to all the members who are part of it, but especially to Eric, Sarah,

Suzy, and all the steering committee members. I had a lot of fun organizing conferences, work in

progress sessions and the JTalks event. I am looking forward to seeing the community grow.

To the Ephys group, it was great to join in many of your lab meetings and events. I know

the distance did not allow us to interact as much as I would have liked, but I hold great memories

of our lab retreat in Barcelona. Thank you for your valuable feedback and interesting

conversations. Marta, thank you for helping me with the in vivo electrophysiology, for hosting me

in your house (even without knowing me), so I did not have to travel so much. I am very happy to

count you as a friend and thank you for introducing me to the Spanish clan!

This is the right time to express my gratitude to all my friends who encouraged and

supported me during these years. When it comes to friends there is an enormous number of people

I would like to thank and I know the chance of accidentally forget someone is rather high. So, let

me start by saying “thank you!” to everyone who in one way or another has been part of my life

in the last 4 years. Steph, you do not have an idea of how much I would like not to have an ocean

between us. I know we do not call each other as much as we should, but every time we do, it feels

like the time has not passed. Steve, Hannes, and Constance thank you for our great annual trips.

Let’s make it to Scotland next year! Thilo & Claartje, thank you for all our weekends and dinners

together. The Spanish group in Leuven, thank you for the amazing dinners que te hacen explotar,

Acknowledgements

and the numerous times I stayed at your apartments. Jime y Marta se han convertido en dos

personas muy importantes en mi vida. Jeroen, I truly enjoy our conversations (especially when we

are drinking one of your amazing beers). Thank you for introducing me to one of the most genuine

people I know, Julie. Thank you to the Tousensemble group for fun weekends together. To the

BMW group for welcoming me into the family with open arms. Stijn and Natascha, I hold great

memories of our trip to Sardinia, I hope we can repeat soon! Constantin and Annelies thank you

for the nice dinners together (where we always talk too much about work) and trusting us with

your kids.

This thank you is one of the most difficult ones because I cannot find the words to express

my gratitude. Louis, you complete me in a way that is inexplicable. Thank you for always listening,

for helping me in the lab, and for giving me scientific advice. We both know how difficult the last

year has been for me. Thank you for being by my side, for always supporting me, for encouraging

me, and for believing in me. I look forward to having many new adventures together. Te amo!

Louis, I also have to thank you for giving me such an amazing “family in law”. Thank you

all for welcoming me into the De Muynck family. Now that I am closing the Ph.D. chapter, I

promise to put more effort into learning Dutch, but I cannot promise I will be able to learn West-

Vlaams…I think this takes a lifetime.

Gracias a mi familia en Argentina, mis hermanos, sobrinos, tías (las de toda la vida y las

mas recientes), primos y primas. Me hubiese gustado que estén aquí, pero entiendo que los 11.279

km no lo permiten. Gracias papa y mama por dejarme volar…mama se que esta orgullosa de mi.

¡Te extraño!

Sofia Jacob

FACULTY OF PSYCHOLOGY & EDUCATIONAL SCIENCES

DEPARTMENT OF BRAIN AND COGNITION

LABORATORY OF BIOLOGICAL PSYCHOLOGY

TIENSESTRAAT 102

3000 LEUVEN, BELGIUM

Neurophysiological Assessment of Brain Network Activity Involved in Cognitive Processing in Animal Models of Alzheimer’s Disease

Sofia Jacob

Neurophysiological Assessment of Brain Network Activity ...

Documents

Transcript of Neurophysiological Assessment of Brain Network Activity ...