Individual performance in cognitive aging and deficit prevention by cognitive training in the rat: integration of structural, molecular and functional correlates
Abstract
O envelhecimento está associado a défices de memória. No entanto, as alterações cognitivas relacionadas com o envelhecimento variam significativamente entre indivíduos e domínios cognitivos. Até ao momento, a evidência sobre os mecanismos por detrás desta heterogeneidade é diminuta. Neste trabalho, fizemos a caracterização comportamental de uma grande coorte de ratos envelhecidos e jovens adultos. Mostramos que os ratos envelhecidos apresentavam uma menor aprendizagem espacial e flexibilidade comportamental. De realçar, o grau de declínio cognitivo foi altamente variável. Dada esta variabilidade, agrupamos os animais envelhecidos e jovens de acordo com a sua performance, caracterizando-os como bons ou maus. De seguida, exploramos os correlatos estruturais, funcionais e moleculares destas diferenças comportamentais no hipocampo e córtex pré-frontal medial, regiões chave envolvidas no funcionamento cognitivo e particularmente vulneráveis ao processo de envelhecimento. Os resultados mostraram que as diferenças individuais da função cognitiva podem ser correlacionadas com alterações estruturais, funcionais e moleculares nestas regiões, no entanto em sentidos opostos, tratando-se de animais novos ou envelhecidos. Enquanto que para os animais novos “maior é melhor”, parece que “menor é melhor” é mais apropriado aos animais envelhecidos. Nestes, a marcada heterogeneidade comportamental poderá ser atribuída a variações nos níveis de neurotrofinas e a um decréscimo da atividade autofágica. Pelo descrito, o desenvolvimento de intervenções que otimizam o funcionamento cognitivo é da maior relevância. Foi já descrito que o treino cognitivo pode aumentar a memória. Por esta razão, ensaiamos se o teste da tábua perfurada seria capaz de prevenir défices cognitivos dependentes de memória. As nossas descobertas sugerem que este treino tem efeitos benéficos duradouros. Potencialmente, o trabalho aqui apresentado poderá contribuir para a compreensão das diferenças cognitivas individuais relacionadas com a idade, podendo auxiliar no desenvolvimento de novas vias terapêuticas para um envelhecimento cerebral saudável.
Full text
Universidade do Minho Escola de Medicina Cristina de Fátima Sousa da Mota julho de 2021 Individual performance in cognitive aging and deficit prevention by cognitive training in the rat: integration of structural, molecular and functional correlates Cristina de Fátima Sousa da Mota Individual performance in cognitive aging and deficit prevention by cognitive training in the rat: integration of structural, molecular and functional correlates UMinho|2021
Cristina de Fátima Sousa da Mota julho de 2021 Individual performance in cognitive aging and deficit prevention by cognitive training in the rat: integration of structural, molecular and functional correlates Trabalho efetuado sob a orientação do Professor Doutor João José Cerqueira e do Professor Doutor João Carlos Sousa Tese de Doutoramento Doutoramento em Ciências da Saúde Universidade do Minho Escola de Medicina
ii DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição 4.0 CC BY https://creativecommons.org/licenses/by/4.0/
iii Agradecimentos/Acknowledgements Ao Professor João Cerqueira por me ter orientado ao longo destes anos e por me ter dado a oportunidade de trabalhar com os meus “velhotes”. Pela sua disponibilidade, por toda a energia e boa disposição características, e pelo agradável “isso” que muitas vezes ouvi quando concordava comigo. Um sincero obrigado! Ao Professor João Sousa por tudo aquilo que me ensinou e, principalmente, por estar sempre lá quando era preciso. Muito obrigada! Ao Grupo das Neurociências por todas as discussões científicas. Aos meus colegas de laboratório pelo apoio, por todos os momentos de boa disposição e pelo companheirismo demonstrado ao longo deste processo. Um agradecimento especial à Ana Rita Marques, Paula Silva, António Melo, Mónica Morais, Susana Monteiro, Sofia Neves, Cláudia Antunes e Marina Amorim. Todos eles tornaram esta aventura muito mais agradável! Aos meus amigos, por sempre me apoiaram. Obrigada por cada momento de boa disposição, por cada aventura e por toda a ajuda! Aos meus pais por toda a compreensão e apoio. Em especial à minha irmã por todos os conselhos, por todo o interesse demonstrado pelos “ratinhos”, por sempre me incentivar e por estares sempre disponível. Em especial, ao Vítor, pela paciência, força e apoio incondicional durante todo este processo, sem ti não teria sido possível. Obrigado por seres quem és! This work was supported by the Portuguese Foundation for Science and Technology (FCT) with a fellowship granted to Cristina Mota (SFRH/BD/81881/2011) and by the European Commission within the 7th framework program, under the grant agreement: Health-F2-2010-259772 (Switchbox). In addition, this work was co-funded by the Northern Portugal Regional Operational Programme (ON.2 SR&TD Integrated Program – NORTE-07-0124-FEDER-000021), through the European Regional Development Fund (FEDER) and by national funds granted by FCT (PEst-C/SAU/LA0026/2013), and FEDER through the COMPETE (FCOMP-01-0124-FEDER-037298).
iv STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho.
v Performance individual no envelhecimento cognitivo e prevenção de défices pelo treino cognitivo no rato: integração de correlatos estruturais, moleculares e funcionais Resumo O envelhecimento está associado a défices de memória. No entanto, as alterações cognitivas relacionadas com o envelhecimento variam significativamente entre indivíduos e domínios cognitivos. Até ao momento, a evidência sobre os mecanismos por detrás desta heterogeneidade é diminuta. Neste trabalho, fizemos a caracterização comportamental de uma grande coorte de ratos envelhecidos e jovens adultos. Mostramos que os ratos envelhecidos apresentavam uma menor aprendizagem espacial e flexibilidade comportamental. De realçar, o grau de declínio cognitivo foi altamente variável. Dada esta variabilidade, agrupamos os animais envelhecidos e jovens de acordo com a sua performance, caracterizando-os como bons ou maus. De seguida, exploramos os correlatos estruturais, funcionais e moleculares destas diferenças comportamentais no hipocampo e córtex pré-frontal medial, regiões chave envolvidas no funcionamento cognitivo e particularmente vulneráveis ao processo de envelhecimento. Os resultados mostraram que as diferenças individuais da função cognitiva podem ser correlacionadas com alterações estruturais, funcionais e moleculares nestas regiões, no entanto em sentidos opostos, tratando-se de animais novos ou envelhecidos. Enquanto que para os animais novos “maior é melhor”, parece que “menor é melhor” é mais apropriado aos animais envelhecidos. Nestes, a marcada heterogeneidade comportamental poderá ser atribuída a variações nos níveis de neurotrofinas e a um decréscimo da atividade autofágica. Pelo descrito, o desenvolvimento de intervenções que otimizam o funcionamento cognitivo é da maior relevância. Foi já descrito que o treino cognitivo pode aumentar a memória. Por esta razão, ensaiamos se o teste da tábua perfurada seria capaz de prevenir défices cognitivos dependentes de memória. As nossas descobertas sugerem que este treino tem efeitos benéficos duradouros. Potencialmente, o trabalho aqui apresentado poderá contribuir para a compreensão das diferenças cognitivas individuais relacionadas com a idade, podendo auxiliar no desenvolvimento de novas vias terapêuticas para um envelhecimento cerebral saudável. Palavras-chave: alterações estruturais, autofagia, défices cognitivos, envelhecimento, heterogeneidade.
vi Individual performance in cognitive aging and deficit prevention by cognitive training in the rat: integration of structural, molecular and functional correlates Abstract Aging is commonly associated with memory impairments. However, age-related changes vary considerably across individuals and cognitive domains. Thus far, the evidence on the mechanisms underlying such heterogeneity is scarce. Thus, in the present work, we behaviorally characterized a very large cohort of old age and young adult male rats. We showed that old rats, on average, had poorer spatial learning and behavioral flexibility than young adults. Of notice, the degree of cognitive decline was highly variable. Given this variability, we clustered young and old animals according to individual cognitive performance and classified them as good and bad performers. We then explored the structural, functional, and molecular correlates of such behavioral differences in the hippocampus and medial prefrontal cortex. Both areas are key regions involved in cognitive functioning and particularly vulnerable to the aging process. The results of our work showed that individual differences in cognitive function can be correlated with structural, functional, and molecular changes in these brain regions, albeit in different directions in young and old animals. While for young individuals “bigger is better”, it seems that “smaller is better” is a more appropriate aphorism in what regards old subjects. Moreover, we provided evidence that, in older animals, dendritic length and volumetric differences, and concomitant behavioral heterogeneity, can be ascribed to variations in neurotrophin levels and decreased autophagic activity. In line with these results, the development of interventions to maintain cognitive functioning is of utmost importance. It’s known that cognitive training can enhance memory function. Therefore, we tested if the Hole Board test was able to prevent age-associated deficits. Our findings suggested that cognitive training had long-lasting improvements in age-induced impairments. Hopefully, the work we herein present will contribute to the understanding of age-related individual differences, both in rodents as in humans. We hope our work will aid in the development of new therapeutic avenues towards a successful brain aging and an increased “mindspan”. Keywords: aging, autophagy , cognitive impairments, heterogeneity, structural correlates.
vii Table of contents AGRADECIMENTOS/ACKNOWLEDGEMENTS ......................................................................................... III STATEMENT OF INTEGRITY ................................................................................................................... IV RESUMO ................................................................................................................................................ V ABSTRACT ............................................................................................................................................ VI LIST OF ABBREVIATIONS ....................................................................................................................... X THESIS PLANNING ................................................................................................................................ XI CHAPTER I ............................................................................................................................................. 1 INTRODUCTION ..................................................................................................................................... 1 1. GENERAL CONSIDERATIONS .......................................................................................................... 2 2. COGNITION AND COGNITIVE FUNCTION ........................................................................................ 3 3. THE HIPPOCAMPUS AND THE PREFRONTAL CORTEX – IMPLICATIONS TO COGNITIVE FUNCTION 7 3.1. HIPPOCAMPUS – STRUCTURAL AND FUNCTIONAL ORGANIZATION AND DEPENDENT BEHAVIORS ................... 7 3.2. PREFRONTAL CORTEX – STRUCTURAL AND FUNCTIONAL ORGANIZATION AND DEPENDENT BEHAVIORS ........ 10 4. THE AGING BRAIN ........................................................................................................................ 12 4.1. NORMAL BRAIN AGING VERSUS PATHOLOGICAL BRAIN AGING .............................................................. 13 4.2. NEUROCOGNITIVE CHANGES IN AGING ........................................................................................... 14 4.3. HETEROGENEITY IN COGNITIVE FUNCTION DURING THE AGING PROCESS ............................................... 16 4.4. STRUCTURAL ALTERATIONS IN THE AGING BRAIN .............................................................................. 17 4.4.1. VOLUMETRIC ALTERATIONS....................................................................................................... 17 4.4.2. DENDRITIC ALTERATIONS ......................................................................................................... 18 4.5. AUTOPHAGY, DENDRITIC PRUNING AND THE AGING BRAIN .................................................................. 22 4.6. THE ROLE OF COGNITIVE TRAINING ON COGNITIVE FUNCTION.............................................................. 23 5. AIMS OF THE STUDY .................................................................................................................... 26 REFERENCES ....................................................................................................................................... 27 CHAPTER II .......................................................................................................................................... 43 STRUCTURAL AND MOLECULAR CORRELATES OF COGNITIVE AGING IN THE RAT ............................... 43 1. ABSTRACT ................................................................................................................................... 45 2. INTRODUCTION ............................................................................................................................ 45 3. RESULTS .................................................................................................................................... 46 3.1. AGE IS ASSOCIATED WITH COGNITIVE DECLINE AND BEHAVIORAL HETEROGENEITY .................................. 46 3.2. AGE TRIGGERS DENDRITIC ATROPHY THAT CORRELATES WITH INDIVIDUAL COGNITIVE PERFORMANCE .......... 48 3.3. IMPAIRED AUTOPHAGY IMPACTS ON DENDRITIC STRUCTURE ............................................................... 52 4. DISCUSSION ................................................................................................................................ 55 5. METHODS ................................................................................................................................... 61
2 INTRODUCTION 1. General considerations Aging is an inevitable biological process characterized by a declining ability to respond to stress, increasing homeostatic imbalance and increased risk of disease eventually ending in death (Tosato et al. , 2007). Nowadays, worldwide, the number of individuals aged over 65 years old is growing faster than any other age group (Lister & Barnes, 2009). As a consequence of cumulative challenges over the lifespan, the majority of these individuals have to deal with alterations in their bodies, including the brain. Thus, this increase in life expectancy has revealed a new “epidemic”: the elderly are at significant risk for dementia, a syndrome characterized by impaired memory and cognitive abilities, affecting their functional independence. Although modern medicine has allowed fixing or replacing many functions on our bodies, our brain is highly customized, and pathological alterations are often irreversible. Hence, for the majority of aged individuals which are in good physical condition, cognitive decline is the main threat to their quality of life. Because of this growing threat, and in line with the recognition that aging is not a disease in itself, but instead comprises natural biological processes that are, nevertheless, amenable to experimental study, the question of the effect of aging upon the brain has received increased attention within the recent years, and the promotion of “mindspan” (the maintenance of mental abilities over the lifespan) has become a top priority (Gallagher et al. , 2011). The development of rational approaches to diagnose and treat unhealthy aging is, therefore, of paramount importance. Accordingly, a thorough understanding of the multiple mechanisms underlying brain aging is of utmost interest. In recent years, researchers have made rapid progress in understanding the neural changes that affect the life course of cognitive capabilities. New animal models have been developed and old ones have become better characterized and standardized (Mitchell et al ., 2015). In addition, advances in brain imaging techniques now permit investigations in aged humans with amazing resolution and sophistication, providing a bridge between human and animal studies. Altogether, these advances will be useful to develop behavioral and technological interventions aiming to maintain cognitive performance in older individuals. Despite all the advances made in the field, one of the main questions still to be answered is the underlying mechanisms that allow some individuals to age without appreciable cognitive decline while others show major losses of cognitive functions. A vast majority of the existing theories report that age-related deficits
3 in cognitive function are progressive and generalized; however, some studies do show heterogeneous patterns of age-related cognitive decline (Santos et al. , 2013). For that reason, the conventional group averaging statistical approach is no longer satisfactory in understanding the true effect of brain aging (Hsieh, 2015). Thus, more than group descriptions and comparisons with younger subjects, it is crucial to consider inter-individual differences and its determinants, often neglected in most of the published studies; in this thesis, as will be shown later, we considered inter-individual differences as central to understand aging. Throughout the aging process, the desire of a long life goes on pair with that of a healthy one. Considering that the longevity is increasing worldwide and that there is a lack of information about the aging process and its heterogeneity, the understanding of the mechanisms that are responsible for age-related cognitive changes becomes increasingly important, in the hope of finding new therapeutic options for limiting the effects of aging on neuronal function. We will next overview key aspects related to aging in the brain, namely cognitive function (Chapter I section 2) and relevant brain structures for cognition (Chapter I section 3); we will also highlight in further detail concepts and processes presently known about the aging brain that are most relevant under the scope of this thesis (Chapter I section 4). 2. Cognition and cognitive function The word "cognition" dates back to the 15th century when it meant "thinking and awareness” (Revlin, 2012). Nowadays, cognition is defined as “those processes by which the sensory input is transformed, reduced, elaborated, stored, recovered, and used” (Neisser, 1967). In general, human cognition refers to the ability to assimilate and process the information that its received from different sources (perception, experience, belief, etc), and to convert it into knowledge (Revlin, 2012). Likewise, Shettleworth in 1998 argued that “animal’s cognition refers to the mechanisms by which they acquire, process, store and act on information from the environment”. Cognition encompasses a variety of cognitive functions, like learning, attention, memory, language, reasoning and decision making (Kandel, 2000). Each of these cognitive functions are responsible for regulation of specific behaviors or actions and are served by more than one neural pathway that has been persistently mapped in the brain (Kandel, 2000). Understanding how these networks produce the cognitive functions of the brain is one of the ultimate challenges of neuroscience. When one functional region, or pathway, is damaged, others may be able to compensate partially for the loss, otherwise, pronounced deficits in specific cognitive functions may emerge. Eventually, this will lead to alterations in
4 behavior that, altogether, are a reflection of the adjusted capacities of the whole functioning brain, rather than the reflex of the loss of damaged brain regions (Kandel, 2000). Nevertheless, this ability to compensate for lost function depends on the affected region and the extension of the injury. As lesion studies have shown, both in humans and in animal models, severe lesions are often accompanied by specific and reproducible cognitive deficits (Squire & Wixted, 2011). In this thesis, we will focus specifically on two major domains of cognitive function: learning and memory. While learning is the process by which the brain acquires information from the external environment, memory is the process by which that knowledge is encoded, stored, consolidated and later retrieved (Martin & Morris, 2002). These processes operate conjointly at the level of neural networks (Holtmaat & Caroni, 2016; Kandel et al. , 2014). The study of human memory skyrocketed from the 1950s on with the work performed on the patient H.M. (Squire & Wixted, 2011). As a result, different kinds of memory and learning were distinguished. Memories can be generally classified according to their content (declarative or nondeclarative), according to their duration (short-term memory or long term-memory), and according to their nature: purely archival (short-term memory or long term-memory) as opposed to working memory, a transient process encompassing both storage and processing functions, the “blackboard of the mind” (Goldman-Rakic, 1996; Squire & Zola, 1996; Rendeiro et al. , 2009; Squire & Wixted, 2011) (Figure 1). Non-declarative (implicit) memory, is rigid, outside of a person’s awareness, and tightly connected to the original stimulus conditions (Squire & Wixted, 2011) (Figure 1). This type of memory is related to abilities (procedural memory) such as the ability to retain memory for motor and cognitive skills (Harada et al. , 2013). It is dependent on the amygdala, cerebellum, striatum and reflex pathways, and its recall is unconscious (Squire, 1986; Squire & Zola, 1996) (Figure 1). This type of memory is relatively stable across the lifespan (Harada et al. , 2013).
5 Figure 1| Taxonomy of memory systems. Declarative and non-declarative memories are the two main types of human memory. Declarative memory is further divided in semantic, episodic and spatial memory, while non-declarative memory comprises associative and non-associative learning, skills and habits (procedural memory). Associative learning occurs through the association of two previously unrelated stimuli, and includes reinforcement, whereas non-associative learning occurs in response to a single stimulus, without reinforcement. Short-term memories are consolidated into long-term memories by repetition (Adapted from Rendeiro et al. , 2009). Unlike nondeclarative memory, declarative memory shows a marked impairment throughout the life course (Harada et al. , 2013). It is highly flexible and represents the factual knowledge of people, places, objects and events, acting to form associations between them; it is the kind of memory that is referred to when the term memory is used in everyday language (Squire & Wixted, 2011). It is supported by the medial temporal lobe and hippocampus (HPC), and its retrieval requires conscious attention (Squire, 1986; Squire & Zola, 1996) (Figure 1). Examples of declarative memory include semantic memory, episodic memory and spatial memory (Figure 1). Semantic memory involves fund of information, language usage, and practical knowledge, for example, knowing the meaning of words (Glisky, 2007; Harada et al. , 2013). Episodic memory refers to the ability of encoding, storing, and consciously recollecting previously learnt events over variable periods ranging from minutes to years (Burgess et al. , 2002; Harada et al. , 2013). It concerns to our ability to consciously recollect personally experienced
6 events including the information related to the ongoing external context such as space, time, as well as who was involved (Burgess et al. , 2002). As humans, rodents acquire declarative memories when they ‘learn’ the required route within a specific maze environment, indicating that they have formed a flexible neural map of the environment to guide their behavior in specific situations (Burgess et al. , 2002). This type of declarative memory, termed spatial memory (Figure 1), is present in both humans and rodents and is commonly divided into two main systems: egocentric (subject-centered), relying mainly on the parietal neocortex, and allocentric (objectcentered), which is dependent mainly on the integrity of the HPC, and is also commonly termed spatial reference memory (Burgess et al. , 2002; Wirt & Hyman, 2017). The anatomy of memory has been a question of continuous debate within the neuroscience community (Jonides et al. , 2008). Particularly, the storage of short-term and long-term memories, as well as the brain regions responsible for the executive processes guiding their integration, are far from being resolved. Nevertheless, the state-based model presently gathers the best consensus. This model assumes that short-term and long term-memories are stored within the same regions of the cortex that where involved in the initial perception and encoding, and that frontal regions are responsible for executive functions (Jonides et al. , 2008; LaRocque et al. , 2014; Nyberg & Erikson, 2015). In this sense, short-term memories, stored in posterior cortical regions, will be maintained and processed following prefrontal cortex (PFC) recruitment, allowing for updating, mnemonic representations of stimuli, manipulation, and interactions with long-term memories, for example by interactions between short-term and long-term memory storages. Therefore, maintenance of novel information in working memory can engage the medial temporal lobe consolidation system, having the HPC as a central participant, leading to the formation of new long-term memories. The medial temporal lobe is thus a highly specialized region of the brain, able to create representations involving novel relations - encoding, while having a pivotal role in memory consolidation and retrieval. The PFC, in turn, has a pivotal role in working memory, functioning as an online maintainer of information, allowing manipulation, updating, coordination of behavior when multiple goals are active, and shifting from one set of rules to another one, the so called behavior flexibility (de Bruin et al. , 1994; Jonides et al. , 2008; LaRoque et al. , 2014; Nyberg & Eriksson, 2015). Declines in declarative memory occur with normal aging, however, the timing of these deficits is different. While episodic and spatial memories show lifelong declines, semantic memory shows late life deterioration (Rendeiro et al. , 2009; Harada et al. , 2013). Thus, an understanding of the biological basis of these cognitive functions, particularly those under the scope of this thesis, learning and memory, requires an appreciation of the anatomy and function of the
7 neural systems that subserve these functions in the brain. Therefore, our work was focused in the study of HPC and PFC circuits, since both are critical structures for learning and memory, and both are involved in the process of brain aging. We will next, refer to these brain regions with further detail. 3. The hippocampus and the prefrontal cortex – implications to cognitive function During the last few years, the emergence of new imaging techniques revolutionized the study of the neuroanatomical basis of cognitive functions. Thus, a much clearer idea about the brain regions involved in many complex cognitive functions is being formed. In this regard, it is now known that the HPC and the PFC are critical structures for learning and memory (Burgess et al., 2002; Squire & Zola, 1996; Damasio, 2000). Because of their crucial role in memory in general, and in working memory and spatial memory, particularly, these two brain structures have been the major focus of a productive line of research, concentrated in the fundamental aspects of cognitive processing. Here, we present a general survey of the literature, regarding the human and rat brain. For each brain region, we first review the structural and functional organization, and then take a deeper look at the dependent behaviors. 3.1. Hippocampus – structural and functional organization and dependent behaviors The HPC is a highly interconnected structure that serves a pivotal role in the formation of declarative memories (Squire, 1992; Morris et al. , 2003), as well as in the regulation of the hippocampal-pituitaryadrenal axis (reviewed by Sousa et al. , 2008). Due to this fundamental role, atrophy in the HPC formation is often an early feature in the progression of Alzheimer’s disease (AD), and patients with damaged HPC often have difficulty in forming new, and enduring memories of personally experienced events. This general episodic amnesia coexists with marked deficits in spatial orientation and navigation (Hartley et al. , 2013). The hippocampal region comprises the HPC formation and the parahippocampal region (Cappaert et al ., 2014). In this thesis we will focus on the HPC formation. The HPC formation is a medial temporal lobe structure belonging to the limbic system. It has a long C-shaped form that is present across all mammalian orders and runs along a dorsal (septal)-to-ventral (temporal) axis in rodents, corresponding to a posterior-to-anterior axis in humans (Moser & Moser, 1998; Strange et al. , 2014). While the dorsal (or posterior) HPC (DHPC) mediates cognitive functions, particularly spatial memory; the ventral (or anterior) HPC (VHPC) is involved in emotional responses (Moser & Moser, 1998; Fanselow & Dong, 2010). In particular, a more dense ventral than dorsal connectivity with the amygdala (Fanselow & Dong,
8 2010), and the selective VHPC role in the endocrine stress response (reviewed by Sousa et al. , 2008), strongly support this dichotomy in HPC cognitive processing. The anatomy of the HPC formation comprises three main subfields, distributed from proximal to distal along the transverse axis of the HPC, with the dentate gyrus (DG) as the most medial and proximal portion, laterally flanked by the cornu ammonis (CA), the HPC proper, with its three subfields (CA1, CA2, CA3), and the subiculum (Sb) (Strange et al ,. 2014). The CA1 and CA3 fields are the HPC regions which were subject to more intense study; in contrast, the smaller CA2 is, so far, the less-studied CA subfield. Indeed, some authors (Lorente de No, 1934) claim that the CA2 subfield consists of CA3-type pyramids which did not receive the mossy fiber contacts coming from the dentate granule cells, while others reported no differences between CA2 and CA3 cells in terms of their connections or histochemical staining properties (Blackstad, 1956). Nevertheless, recent studies have revealed unique properties and an influential role of this region in encoding social, temporal and contextual aspects of memory (Robert et al. , 2018). The aforementioned structures constitute the HPC network, which is primarily a unidirectional network. The HPC receives major inputs from the entorhinal cortex (EC), which projects to the granule cells of the DG and CA3 pyramidal neurons via the perforant path. The axons of the granule cells are termed mossy fibers and send inputs mainly to the apical dendrites of the pyramidal cells of the HPC area CA3. CA3 neurons, in their turn, send axons to the apical dendrites of CA1 pyramidal cells via the Schaffer collateral pathway, as well as to CA1 cells in the contralateral HPC via the associational commissural pathway. CA1 neurons also receive input directly from the perforant path; and they are the last point of this trisynaptic circuit, sending axons to the Sb. Subicular neurons close this circuit by sending backprojections to the EC, forming, therefore, a reverberating loop (Strange et al. , 2014; Kesner & Rolls, 2015). This basic intrinsic circuitry is maintained throughout the long axis and across species. Also, a cross-species comparison of anatomical connectivity provides evidence that the primate HPC long axis may be homologous to that of the rat (Strange et al. , 2014). The present thesis will focus on the study of the DG, CA3 and CA1. Although the role of the HPC is most often evaluated as a single entity, recent evidence from a variety of experimental approaches including lesion, behavioral, electrophysiological and gene activation studies, suggests that there are separable and distinct sub-regional functions within this structure (Kesner et al. , 2004). It is known that the DG processes metric spatial representation, performing complex pattern separation computations, while the CA3 subregion is responsible for spatial pattern association and completion, detection of novelty and shortterm memory. The CA1 subregion mediates processes involved in temporal pattern association and
9 completion as well as intermediate-term memory, functioning as an important output hub strongly involved in memory consolidation (Kesner et al. , 2004). Furthermore, given the prominent role of the HPC in memory, it is no surprise that the HPC and PFC are anatomically related. Indeed, they are connected through the projection of axons originating in the Sb and ventral CA1 subfields, which terminate in the pyramidal cells and interneurons of the medial PFC (mPFC) (Tierney et al. , 2004) (Figure 2). The pathway is unidirectional but may be reciprocated via a bi-synaptic route through the nucleus reunions or lateral EC (Vertes et al. , 2007) (Figure 2). Similarly, projections arising from the mPFC are returned to the HPC through a direct connection from mPFC to dorsal CA1 or via the EC through the nucleus reuniens and also via the medial dorsal thalamic nuclei (Wirt & Hyman, 2017) (Figure 2). The integrity of the mPFC-HPC connection is necessary for spatial working memory in paradigms including the water maze (Wang & Cai, 2008), the T-maze (Wang & Cai, 2006) and spatial win-shift on the radial arm maze (Goto & Grace, 2008). Modulation of synaptic activity in the connection between these two areas contributes to a synergistic regulation of learning/memory processes (Cerqueira et al. , 2007a). Summarizing, due to the multiple and disparate functions that have been ascribed to the HPC, it has been intensively investigated in the human and rat brain in the context of understanding memory, learning processes, neurodegenerative diseases and cognitive decline. Figure 2| Connections between the HPC and the mPFC. Projecting axons originating in the Sb and ventral CA1 subfields terminate in the pyramidal cells and interneurons of the medial mPFC (for more details see text). medial prefrontal cortex = mPFC; nucleus reuniens = NR; mediodorsal thalamic nuclei = MDT; subiculum = Sb; CA1 = dorsal (top) ventral (bottom). (From Wirt & Hyman, 2017).
10 3.2. Prefrontal cortex – structural and functional organization and dependent behaviors The PFC is critical to many cognitive abilities that are considered particularly human, and is part of a large neuronal system crucial for normal affective behavior and executive functioning, both in humans and other primates (Teffer & Semendeferi, 2012). In general, the PFC plays a key role in decision making (Manes et al. , 2002), planning (Muller et al. , 2002), processing of emotional stimuli and behavioral flexibility, as well as in social interactions (Damasio, 2000). Also, as already discussed, the PFC has a central role in working memory (Damasio, 2000), which involves transient storage and manipulation of information to guide subsequent behavior. For a long time the PFC was thought unique to the primate species and called the “frontal granular cortex” (Uylings et al. , 2003). It comprises a group of cortical areas which are structurally and functionally heterogeneous (Lewis, 2004). It includes all cortical areas of the primate frontal lobe that have an inner granular layer IV and lie rostral to the agranular (pre)motor region. In primates, these areas can be roughly divided into different anatomic subfields namely a dorsolateral, a medial (anterior cingulate) and an orbital region (Uylings et al. , 2003; Lewis, 2004). Damage to the human dorsolateral frontal region is characterized especially by deficits in working memory. Damage to the orbitofrontal region is characterized by altered socio-emotional behaviors, hyperkinesis, deficits in the processing of olfactory and gustatory information, and in spontaneity. The medial (anterior cingulate) region is not as well characterized, but its function is related to attentional processes related to internal states. Thus, damage to anterior cingulate regions can include reduced response to pain, akinetic mutism, and impaired motor initiation (Uylings et al. , 2003). While most authors agree on this anatomical distinction, they differ in the different functions that have been ascribed to these subdivisions of the primate PFC, particularly along its dorsal-ventral axis (GoldmanRakic, 1995; Damasio, 2000). For instance, regarding the contributions of the different PFC subareas to working memory, two principal theories have emerged - the ‘‘type of information’’ theory and the ‘‘type of processing’’ necessary for task completion (Goldman-Rakic, 1995; Muller et al. , 2002). In the first theory, the primate dorsolateral PFC is ascribed to spatial memory content whereas more ventral areas subserve memory for objects. On the other hand, in the “type of processing’’ necessary for task completion theory, PFC ventrolateral regions would subserve active maintenance of both object and spatial information within memory, whereas the dorsolateral region is required for ‘‘monitoring’’ and ‘‘manipulation’’ of working memory information (Muller et al. , 2002). The volume of the cerebral cortex of a rat is about a hundred times smaller than that of the cerebral cortex of macaques, and about a thousand times smaller than that of humans. This decreased volume is paralleled by less evolution, less differentiation, and less segregation than the corresponding primate
11 cerebral cortex. This fact has raised controversy whether or not rats possess a prefrontal region comparable with the primate PFC. Nevertheless, the use of other classification aspects rather than the sole use of cytoarchitectonic criteria has progressively put aside these early disagreements about the existence of a PFC in non-primates. Thus, it is now generally accepted that connectivity, functionality, neurotransmitter milieu and embryological development should be considered, together with cytoarchitectonic characteristics, when discussing homologies between cortical areas in different species (Uylings et al. , 2003). Taking these criteria under consideration, one can delineate a region on the frontal pole of the rat brain that may be considered equivalent to the primate PFC (Uylings et al. , 2003). The rodent PFC can be grossly divided into two main regions: a medial region (mPFC), that has characteristics of both dorsolateral and medial primate subdivisions, and a lateral and ventral subfield, equivalent to the primate orbital region (OFC) (Heidbreder & Groenewegen, 2003; Dalley et al. , 2004; Zilles & Wree, 2004). The rat mPFC, as a whole, has been traditionally implicated in attentional processes, working memory and behavioral flexibility, therefore, throughout the rest of this thesis, together with the HPC, this brain region will be the focus of our study. The rodent mPFC receives diverse afferent inputs from limbic regions, including the amygdala and HPC (CA1 and Sb), and provides direct outputs to hypothalamic and numerous brainstem areas involved in the regulation of emotion and the physiological response to stress (Bandler et al. , 2000). It can be further divided into the frontal area 2, dorsal and ventral anterior cingulate areas (Cg), prelimbic area (PL), infralimbic area (IL) and medial orbital area (MO) (Van Eden, 1985). Furthermore, there is now consensual agreement on a main subdivision of the mPFC into dorsal (mainly Cg and PL) and ventral (mainly IL) components. While ventral regions are specialized for autonomic/emotional control, dorsal regions regulate working memory and some forms of motor sequencing (Uylings et al. , 2003; Heidbreder & Groenewegen, 2003). Thus, the anterior Cg cortex and the dorsal PL area have been implicated in attention and working memory (Cerqueira et al. , 2008; Euston et al. , 2012; Heidbreder & Groenewegen, 2003); while the ventral IL area has been associated to goal-directed behavior and autonomic functions (Heidbreder & Groenewegen, 2003; Euston et al. , 2012). Both PL and IL areas have been associated with the performance in behavioral flexibility tasks (Ragozzino et al. , 1999). In summary, the rat PFC is subdivided into a ventrolateral area (OFC area), that plays a central role in the control of socio-affective behaviors; a dorsomedial area (mainly Cg and PL) that regulates working memory, and a ventromedial (mainly IL area) that is involved in visceromotor behaviors. The present
18 already shown that volumetric alterations in areas implicated in cognitive abilities could be an important determinant for elderly cognitive function (Driscoll et al. , 2003, 2006; reviewed by Dickstein et al. , 2007). It is also known that specific brain regions are more susceptible to shrinkage than others, and that changes in brain volume are not uniform across areas/individuals (Raz et al. , 2005). In this regard, the HPC, frontal areas and striatum, are among the regions most affected by age in healthy humans (Raz et al. , 2005; Peters, 2006). Historically, it has been presumed that brain volume loss in HPC and mPFC areas underpin age-related cognitive decline, in both humans (Jernigan et al. , 1991; Golomb et al. , 1993 Driscoll et al. , 2003; 2009; Freeman et al. , 2008; Raz et al. , 2010) and rats (Rapp et al. , 1999; Driscoll et al. , 2006; Yates et al. , 2008). While HPC atrophy has been associated with deficits in episodic and spatial memory (Golomb et al. , 1993; Van Petten, 2004), PFC atrophy is associated with deficits in executive function and working memory (Cerqueira et al. ,2005). The implicit link for aging is, of course, an existence of volumetric atrophy which impairs function. However, despite the several studies already addressing volume loss and cognitive deficits during the aging process, contradictory findings constantly challenge this common assumption (Sullivan et al. , 1995; Raz, 1996; Van Petten, 2004), which could be justified by multiple factors that affect the course of brain aging. For instance, Raz et al. (2003) showed that hypertension exacerbates age-related shrinkage in prefrontal regions; Coffey et al. (1998) reported that women show lesser brain aging than men; Satz (1993) demonstrated that larger brain volume is a neuroprotective factor, and other studies reported that higher formal education delays brain aging and ameliorates its course (Stern et al. , 1992; Kramer et al. , 2004). Likewise, it is important to highlight that the heterogeneity observed in the aging brain has not been a focus of attention in previous aging studies, which could possibly represent a significant confounding factor. Given the above, the experimental work presented in this thesis (Chapter III ) was carried bearing in mind the possible relation between normal aging heterogeneity and the consequent brain atrophy and cognitive decline, with a bias towards the mPFC and HPC, as these brain regions were shown to be the most commonly affected during the aging process. Moreover, as there is evidence of an existing relationship between neuronal remodeling and volumetric alterations (Cerqueira et al. , 2005, 2007b), we also tackled the relation between neuronal dendritic branching and cognitive performance. 4.4.2. Dendritic alterations Neuronal dendritic trees are important in the formation and maintenance of neural networks, regulation of synaptic plasticity and the integration of electrical inputs (Jan & Jan, 2010). Therefore, alterations on their structure impact on cognitive function. More importantly, changes in their morphology were shown
19 to correlate with cognitive alterations in both rodents and humans (see as an example Becker et al. , 1986 and Cerqueira et al. , 2005). In this regard, a large body of literature has examined the state of neurons during aging and has shown contradictory results. Some studies reported an age-related regression in the dendritic arbors of the neuronal populations in the HPC and mPFC, while others showed the opposite or even an absence of dendritic remodelling (see table 1 and 2 for more details). Therefore, similarly to the conclusions redrawn from volumetric studies, several studies have also reported age-related cognitive decline to be associated with dendritic atrophy in areas implicated in cognitive abilities, such as the HPC and the mPFC (Driscoll et al. , 2003, 2006; de Brabander et al. , 1998; Markham & Juraska, 2002; Dickstein, 2013). However, as before, apparent contradictions between studies undermine the establishment of a general consensus.
20 Table 1: Morphological alterations in HPC neurons during the aging process.
21 Table 2: Morphological alterations in mPFC neurons during the aging process.
22 Nevertheless, taken together, these age-related morphological changes seem to be selective, although without a universal pattern across the entire brain. This general lack of agreement in findings could be ascribed to the critical forgetfulness of the heterogeneity observed in the aging brain. In sum, several studies reported that atrophy of dendritic branching and volume loss in the HPC and mPFC underpin age-related cognitive decline (de Brabander et al. , 1998; Markham & Juraska, 2002; Driscoll et al. , 2003, 2006; reviewed by Disckstein, 2013). However, despite these reports of age-related structural variations, little is known regarding the relationship between these variations and age-related decline in HPC and mPFC-dependent learning and memory. Also, even acknowledging the evidence for an age-associated decline of cognitive functioning, on average, one cannot oversee the fact that some aged individuals, both humans and rodents (Gallagher et al. , 2003; Ardila, 2007; Ménard & Quirion, 2012; Santos et al. , 2013), seem to perform as well as their young counterparts. Therefore, in the present thesis, we decided to tackle this question and explore cognitive aging and its structural correlates from an individual perspective in a very large group of aged and young male Wistar Han rats. By addressing inter-individual variation within each age group we expected to tackle its underpinnings and identify some of its determinants. 4.5. Autophagy, dendritic pruning and the aging brain Given the described above (section 4.4) on the dendritic alterations that occur during aging, it is worth of mention that in the developing brain, the formation of dendritic trees is followed by selective pruning (Puram et al. , 2011). Hence, the homeostasis of the mammalian neuroarchitecture is a dynamic process involving a balance between synaptic sprouting and pruning. The mechanisms underlying these processes are particularly active during development and pathological neurodegeneration but are also functional in physiological conditions (Wong & Ghosh, 2002; Querfurth & LaFerla, 2010;Tang et al. , 2014); one such example is the activity-dependent remodeling of synapses and dendritic trees, vital for multiple brain functions including learning and memory (Wong & Ghosh, 2002; Yin & Yuan, 2015; Sugie et al. , 2018). Of note, under normal circumstances, the relative importance between dendritic sprouting and pruning varies throughout the lifespan, with synapse and dendritic formation generally exceeding pruning during brain development, while an opposite trend is observed in the adult brain (Puram et al. , 2011;Tang et al. , 2014). Curiously, a review from Van Petten, 2004, interestingly discusses that the loss of cortical gray matter during childhood and adolescence is responsible for an improvement in cognitive abilities. On the other hand, an excessive dendritic arbor is thought to be associated with neurodevelopmental disorders such as autism spectrum disorders (ASD) (Tang et al. , 2014; Kim et al. ,
23 2017). Taken together, these evidences highlight the fact that, as for the previously described contradictions on age-dependent morphological remodeling, the significance of the balance between sprouting and pruning is still largely unknown. The maintenance of sprouting and pruning balance is under tight control through protein synthesis and autophagic recycling (Bingol & Shang, 2011; Puram et al ., 2011; Tang et al. , 2014; Kanamori et al. , 2015). Autophagy, a lysosome-dependent degradation mechanism, is thus as an important catabolic process that, by degrading long-lived synaptic proteins and damaged organelles, importantly impacts on dendritic turnover and synaptic function (Rubinsztein et al. , 2011; Shehata & Inokuchi, 2014; Gupta et al. , 2016). Therefore, synaptic pruning has a critical role in maintaining CNS homeostasis through the modulation of neuronal density, synaptic remodeling and structural plasticity. Thus, in parallel to the study of classical strategies to prevent cognitive decline, such as cognitive training, which we will explore in the following section (section 4.6) of this introduction, our work was also directed towards the study of alternative approaches such as the autophagy system. Our results (Chapter II) undoubtedly support that the modulation of the autophagy molecular avenue is a promising target for future age therapy interventions. 4.6. The role of cognitive training on cognitive function Since memory impairments impose a substantial burden for the affected individuals, the development of interventions to maintain and optimize cognitive functioning is of utmost importance. Several studies have tried to tackle on strategies to prevent cognitive decline in old age, suggesting interventions to achieve successful cognitive aging based on: i) nutritional improvements (Richards et al. , 2002), ii) physical exercise (Tanigawa et al. , 2014), iii) participation in certain activities - the lifestylecognition hypothesis (Marioni et al. , 2012), building cognitive reserve (Stern, 2002) and cognitive training (Willis et al. , 2006; Nouchi et al. , 2012; Jiang et al. , 2016). However, the progress in this field is slow and there are essentially no interventions that improve memory reliably. In humans, cognitive training, both early and later in life, through engagement in intellectually stimulating activities, is a powerful modulator of the brain and is associated with better cognitive functioning, as it postpones or attenuates cognitive decline, and ameliorates cognitive deficits (Wilson et al. , 2002; Frick & Benoit, 2010; Milgram et al. , 2006; Belleville & Bherer, 2012; Rebok et al. , 2014). Impressively, these improvements can be maintained for years (Willis et al. , 2006). For example, in the ACTIVE trial, a randomized multicenter trial involving cognitively normal older adults, cognitive training resulted in
24 improved cognitive abilities specific to the abilities trained that continued 5 years after the initiation of the intervention (Willis et al. , 2006). Thus, there is sufficient evidence suggesting any activity which involves thinking and learning is beneficial when it comes to maintaining or improving brain health and preventing dementia. The more complex and challenging the mental activities, the greater benefit they offer. Also, the more brain activities and the higher frequency of enrolment, the lower is the risk for dementia. In this thesis, we will address the effects of brain tasks that can help improve and/or maintain brain health. The effects of cognitive training could be explained by the cognitive reserve hypothesis, which rests on the ability of the brain to compensate for pathological changes associated with aging, depending on the previous stage of intellectual capability (Whalley et al. , 2004; Stern, 2002). The concept of cognitive reserve originated in the late 1980s, when a study performed by Katzman et al. , 1988 reported that some individuals with brain changes compatible with Alzheimer's disease, clinically presented none or very little symptoms of dementia. The investigators speculated that these individuals did not show symptoms of the disease while they were alive because they started with larger brains and more neurons, and thus might be said to have had a greater "reserve" to the offset of damage and, therefore, continued to function as usual. While passive reserve refers to brain reserves such as brain volume and the number of neurons and synapses, active reserve corresponds to the brain’s potential for plasticity and reorganization of neuronal networks, allowing it to cope with or compensate for pathology (Stern, 2002; Harada et al. , 2013). This hypothesis supports the idea that higher levels of education, engagement in certain activities, higher socioeconomic status, and baseline intellectual capability, protect against the neuroanatomical alterations observed during the aging process (Stern, 2002). Furthermore, the concept of cognitive reserve has been put forward to account for individual differences in susceptibility to age-related brain changes (Stern, 2012). Accordingly, in humans, cognitive training induces increased dendritic length of neurons (Jacobs et al. , 1993), increased volumes in the HPC (Maguire et al. , 2006; Cannonieri et al. , 2007), cortical and subcortical areas (Seider et al. , 2016), and resulted in important changes in brain activity (Hempel et al. , 2004; Olesen et al. , 2004). In rodents, the beneficial effects of cognitive training have also been associated with physical stimulation, which together is often referred to as environmental enrichment (EE) (Fischer, 2016). Environmental enriched rodents are typically socially housed in large groups and exposed to a variety of stimulating objects that can provide both cognitive stimulation (e.g., toys, tunnels, dwellings) and physical exercise (e.g., running wheels). For example, Vicens et al. , (1999, 2002) found that a water maze training task,
25 performed in 6-month-old mice, led to an improved performance in the water maze later at 10 and 18 months of age. More recently, Galeano et al. (2015) showed that life-long EE was able to rescue memory deficits in control aged rats and in aged rats that were subjected to asphyxia at birth. Additionally, several research groups have systematically shown that environmental enrichment induces synaptogenesis (Rampon et al. , 2000), neurogenesis (Kempermann et al. , 2002), cortical thickening (Mohammed et al. , 2002) and dendritic branching (Faherty et al. , 2003; Kolb et al. , 2003; Bindu et al. , 2007). Interestingly, these anatomical changes correlate with functional alterations such as increased hippocampal long-term potentiation (LTP) (Artola et al. , 2006, Stein et al . 2016). Finally, some studies assessing the physiological responses to enrichment reported variations in corticosterone levels (Schrijver et al. , 2002; MorleyFletcher et al. , 2003; Belz et al. , 2003; Moncek et al. , 2004; Konkle et al. , 2010). Despite all the abovementioned evidences, little is known about EE effects in aging animals. Previous unpublished work in our lab has shown that in young adult rats, short cognitive training in a hole-board (HB) or the T-maze (TM) task could effectively and rapidly promote recovery from stressinduced deficits, by triggering structural, electrophysiological and ultimately functional plasticity. Importantly, this previous work also showed that, while training in the HB resulted in improved plasticity in hippocampal dependent networks and spatial memory, training in the TM improved plasticity in PFC networks and executive function, in a highly selective manner. In particular, given the heterogeneity described above, it remains to be shown if all old individuals benefit from cognitive training, or if some could perceive it as a stressor (Mohapel et al. , 2006), given the increased cognitive load imposed on already fragile cognitive abilities. As already discussed, the HPC is one of the brain areas particularly vulnerable to aging (Mora et al. , 2007). Therefore, in this work, and bearing in mind that a specific cognitive stimulation could differently and specifically enhance a particular brain circuit, an HB food-retrieval task, designed to engage spatial reference memory (van der Staay et al. , 2012), was used as a cognitive-training tool in order to enhance HPC-dependent cognitive function (Van der Staay et al. , 1999, 2012; Depoortère et al. , 2010). Considering the increased number of people over the age of 65 and the intense promotion of “cognitive exercises” as a panacea for older persons, these answers are increasingly relevant. Clearer knowledge on what delays cognitive decline and how we can maximize cognitive function is crucial to improve quality of life for these individuals.
26 5. Aims of the study As the number of people over the age of 65 are increasing, an aged population is a challenge that our societies have to face. As discussed above, the cumulative effects over the lifespan, induce several structural, molecular and functional changes in the brain. Although modern medicine can fix or replace many malfunctions on our bodies, our brains are highly customized and irreversible. Therefore, the development of new therapeutic tools to tackle the maladaptive changes imposed by aging is of immediate importance. Understanding the mechanisms of aging is thus vital for this modern endeavor. Under normal aging conditions, there is a significant variability in age-related interindividual cognitive changes, with some individuals displaying age-related cognitive decline, while others present none or minimal age-related dysfunctions. Understanding why some individuals can retain their “youth brains” while others lose cognitive ability, might be vital in an aging society. So far, evidence on the mechanisms underlying such heterogeneity are scarce. Additionally, since memory impairments impose a substantial burden for those affected, the development of interventions to maintain and optimize cognitive functioning is of utmost importance. Hence, cognitive enrichment/training has been associated with better cognitive functioning, as it postpones or attenuates cognitive decline, and ameliorates cognitive deficits. However, current knowledge has not yet established an explicit relationship between age-related cognitive decline, cognitive training, and the remodeling of brain structures that underlie these processes. In order to address some of the important questions that arose from human studies of aging, in this project we proposed a multi-level analysis in the rat, to explore brain aging and its modulation by cognitive training. The present work aimed to: ▪ Cluster both aged and young animals by their cognitive performance in behavioral assessment as cognitively intact (good performers) and cognitively impaired (bad performers) and look for structural (dendritic morphology and volumetric alterations in specific brain regions) and molecular correlates of such differences (Chapters II and III). ▪ Clarify the impact of cognitive training during adulthood on the patterns of cognitive aging and its structural correlates, acknowledging the well described heterogeneity of brain aging (Chapter IV).
27 References • Aggleton, J.P., Blindt, H. S. & Candy, J. M. Working memory in aged rats. Behav. Neurosci. 103, 975-983 (1989). • Allard, S., Scardochio, T., Cuello, & Ribeiro-da-Silva, A.C. Correlation of cognitive performance and morphological changes in neocortical pyramidal neurons in aging. Neurobiol. Aging. 33, 1466-1480 (2012). • Anderson, R.M., Birnie, A.K., Koblesky, N.K., Romig-Martin, S.A. & Radley, J.J. Adrenocortical status predicts the degree of age-related deficits in prefrontal structural plasticity and working memory. J. Neurosci. 34, 8387-8397 (2014). • Ardila, A. Normal aging increases cognitive heterogeneity: analysis of dispersion in WAIS-III scores across age. Arch. Clin. Neuropsychol. 22, 1003-1011 (2003). • Artola, A., von Frijtag, J.C., Fermont, P.C., Gispen, W.H., Schrama, L.H., Kamal, A. & Spruijt, B.M. Long-lasting modulation of the induction of LTD and LTP in rat hippocampal CA1 by behavioural stress and environmental enrichment. Eur. J. Neurosci. 23, 261-272 (2006). • Bäckman, L. Memory and cognition in preclinical dementia: what we know and what we do not know. Can. J. Psychiatry. 53, 354-360 (2008). • Bandler, R., Keay, K.A., Floyd, N. & Price, J. Central circuits mediating patterned autonomic activity during active vs.passive emotional coping. Brain Res. Bull. 53, 95-104 (2000). • Barrera, A., Jiménez, L., González, G.M., Montiel, J. & Aboitiz, F. Dendritic structure of single hippocampal neurons according to sex and hemisphere of origin in middle-aged and elderly human objects. Brain Res. 906, 31-37 (2001). • Barrett, G. L., Bennie, A., Trieu, J., Ping, S. & Tesafoulis, C. The chronology of age-related spatial learning impairment in two rat strains, as tested by the Barnes maze. Behav. Neurosci. 123, 533538 (2009). • Becker, L.E., Armstrong, D.L. & Chan, F. Dendritic atrophy in children with Down's syndrome. Ann. Neurol. 20, 520-526 (1986). • Belleville, S. & Bherer, L. Biomarkers of cognitive training effects in aging. Curr. Transl. Geriatr. Exp. Gerontol. Rep. 1, 104-110 (2012). • Belz, E.E., Kennell, J.S., Czambel, K., Rubin, R.T. & Rhodes, M.E. Environmental enrichment lowers stress-responsive hormones in singly housed male and female rats. Pharmacol. Biochem. Behav. 76, 481-486 (2003).
34 • Kesner, R.P., Lee, I. & Gilbert, P. A behavioral assessment of hippocampal function based on a subregional analysis. Rev. Neurosci. 15, 333-351 (2004). • Kim, H.J., Cho, M.H., Shim, W.H., Kim, J.K., Jeon, E.Y., Kim, D.H. & Yoon, S.Y. Deficient autophagy in microglia impairs synaptic pruning and causes social behavioral defects. Mol. Psychiatry. 22, 1576-1584 (2017). • Kolb, B., Gorny, G., Söderpalm, A.H. & Robinson, T.E. Environmental complexity has different effects on the structure of neurons in the prefrontal cortex versus parietal cortex or nucleus accumbens. Synapse. 48, 149-153 (2003). • Konkle, A.T.M., Kentner, A.C., Baker, S.L., Stewart, A. & Bielajew, C. Environmental-enrichmentrelated variations in behavioral, biochemical and physiological response of Sprague-Dawley and Long Evans rats. J. Am. Assoc. Lab. Anim. Sci. 49, 427-436 (2010). • Kougias, D.G. Nolan, S.O., Koss, W.A., Kim, T., Hankosky, E.R., Gulley, J.M. & Juraska, J.M . Betahydroxy-beta-methylbutyrate ameliorates aging effects in the dendritic tree of pyramidal neurons in the medial prefrontal cortex of both male and female rats. Neurobiol. Aging. 40, 78-85 (2016). • Kramer, A.F., Bherer, L., Colcombe, S.J., Dong, W. & Greenough, W.T. Environmental influences on cognitive and brain plasticity during aging. J. Gerontol. A. Biol. Sci. Med. Sci. 59, 940-957 (2004). • LaRocque, J.J., Lewis-Peacock, J.A. & Postle, B.R. Multiple neural states of representation in shortterm memory? It’s a matter of attention. Front. Hum. Neurosci . 8, 5 (2014). • Lewis, D.A. Structure of the human prefrontal cortex. Am. J. Psychiatry. 161, 1366 (2004). • Lister, J.P. & Barnes, C.A. Neurobiological changes in the hippocampus during normative aging. Arch. Neurol. 66, 829-833 (2009). • Lolova, I. Dendritic changes in the hippocampus of aged rats. Acta Morphol. Hung. 37, 3-10 (1989). • Lorente de No, R. Studies on the structure of the cerebral cortex. II. Continuation of the study of the Ammonic system. J. Psychol. Neurol . 46, 113-177 (1934). • Luebke, J.I. & Rosene, D.L. Aging alters dendritic morphology, input resistance, and inhibitory signaling in dentate granule cells of the rhesus monkey. J. Comp. Neurol. 460, 573-584 (2003). • Luine, V. & Hearns, M. Spatial memory deficits in aged rats: contributions of the cholinergic system assessed by ChAT. Brain Res. 523, 321-324 (1990). • Lupien, S., Lecours, A.R., Lussier, I., Schwartz, G., Nair, N.P. & Meaney, M.J. Basal cortisol levels and cognitive deficits in human aging. J. Neurosci. 14, 2893-2903 (1994). • Machado-Salas, J.P. & Scheibel, A.B. Limbic system of the aged mouse. Exp. Neurol. 63, 347-355 (1979).
35 • Maguire, E.A., Woollett, K. & Spiers, H.J. London taxi drivers and bus drivers: a structural MRI and neuropsychological analysis. Hippocampus. 16, 1091-1101 (2006). • Manes, F., Sahakian, B., Clark, L., Rogers, R., Antoun, N., Aitken, M. & Robbins, T. Decision-making processes following damage to the prefrontal cortex. Brain. 125, 624-639 (2002). • Marioni, R.E., van den Hout, A., Valenzuela, M.J., Brayne, C., Matthews, F.E., MRC Cognitive Function & Ageing Study. Active cognitive lifestyle associates with cognitive recovery and a reduced risk of cognitive decline. J. Alzheimers Dis. 28, 223-230 (2012). • Markham, J.A. & Juraska, J.M. Aging and sex influence the anatomy of the rat anterior cingulate cortex. Neurobiol. Aging. 23, 579-588 (2002). • Markham, J.A., McKian, K.P., Stroup, T.S. & Juraska, J.M. Sexually dimorphic aging of dendritic morphology in CA1 of hippocampus. Hippocampus. 15, 97-103 (2005). • Martin, S.J. & Morris, R.G.M. New life in an old idea: the synaptic plasticity and memory hypothesis revisited. Hippocampus. 12, 609-636 (2002). • McEwen, B.S. & Seeman, T. Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann. N. Y. Acad. Sci. 896, 30-47 (1999). • McQuail, J.A. & Nicolle, M.M. Spatial reference memory in normal aging Fischer 344 x Brown Norway F1 hybrid rats. Neurobiol. Aging. 36, 323-333 (2015). • Ménard, C. & Quirion, R. Successful cognitive aging in rats: a role for mGluR5 glutamate receptors, homer 1 proteins and downstream signaling pathways. PLoS One. 7, 28666; 10.1371/journal.pone.0028666 (2012). • Mervis, R. Structural alterations in neurons of aged canine neocortex: a golgi study. Exp. Neurol. 62, 417-432 (1978). • Milgram, N.W., Siwak-Tapp, C.T., Araujo, J. & Head, E. Neuroprotective effects of cognitive enrichment. Ageing Res. Rev. 5, 354-369 (2006). • Mitchell, S.J., Scheibye-Knudsen, M., Longo, D.L. & de Cabo, R. Animal models of aging research: implications for human aging and age-related diseases. Annu. Rev. Anim. Biosci. 3, 283-303 (2015). • Mohammed, A. H., Zhu, S. W., Darmopil, S., Hjerling-Lefler, J., Ernfors, P., Winblad, B., Diamond, M.C., Eriksson, P.S. & Bogdanovic, N. Environmental enrichment and the brain. Prog. Brain Res. 138, 109-133 (2002). • Mohapel, P., Mundt-Petersen, K., Brundin, P. & Frielingsdorf, H. Working memory training decreases hippocampal neurogenesis. Neuroscience. 142, 609-613 (2006).
36 • Moncek, F., Duncko, B., Johansson, B. & Zezova, D. Effect of environmental enrichment on stress related systems in rats. J. Neuroendocrinol. 16, 423-431 (2004). • Mora, F., Segovia, G. & Del Arco, A. Aging, plasticity and environmental enrichment: structural changes and neurotransmitter dynamics in several areas of the brain. Brain Res. Rev. 55, 78-88 (2007). • Morley-Fletcher, S., Rea, M., Maccari, S. & Laviola, G. Environmental enrichment during adolescence reverses the effects of prenatal stress on play behaviour and HPA axis reactivity in rats. Eur. J. Neurosci. 18, 3367-3374 (2003). • Morris, R. Developments of a water-maze procedure for studying spatial learning in the rat. J. Neurosci. Methods. 11, 47-60 (1984). • Morris, R.G., Moser, E.I., Riedel, G., Martin, S.J., Sandin, J., Day, M. & O'Carroll, C. Elements of a neurobiological theory of the hippocampus: the role of activity-dependent synaptic plasticity in memory. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 358, 773-786 (2003). • Moser, M.B. & Moser, E.I. Functional differentiation in the hippocampus. Hippocampus. 8, 608-619 (1998). • Muller, N.G., Machado, L. & Knight, R.T. Contributions of subregions of the prefrontal cortex to working memory: evidence from brain lesions in humans. J. Cogn. Neurosci. 14, 673-686 (2002). • Nakamura, S., Akiguchi, I., Kameyama, M. & Mizuno, N. Age-related changes of pyramidal cell basal dendrites in layers III and V of human motor cortex: a quantitative Golgi study. Acta Neuropathol. 65, 281-284 (1985). • Neisser, U. Cognitive Psychology . (ed. Neisser, U.) 4 (Meredith, 1967). • Nichols, N.R., Zieba, M. & Bye, N. Do glucocorticoids contribute to brain aging? Brain Res. Rev. 37, 273-286 (2001). • Nouchi, R., Taki, Y., Takeuchi, H., Hashizume, H., Akitsuki, Y., Shigemune, Y., Sekiguchi, A., Kotozaki, Y., Tsukiura, T., Yomogida, Y. & Kawashima, R. Brain training game improves executive functions and processing speed in the elderly: a randomized controlled trial. PloS One. 7, 29676; 10.1371/journal.pone.0029676 (2012). • Nyberg, L. & Eriksson, J. Working memory: maintenance, updating, and the realization of intentions. Cold Spring Harb. Perspect. Biol. 8, 021816; 10.1101/cshperspect.a021816 (2015). • Nyberg, L., Lovden, M., Riklund, K., Lindenberger, U. & Backman, L. Memory aging and brain maintenance. Trends Cogn. Sci. 16, 292-305 (2012).
37 • Olesen, P.J., Westerberg, H. & Klingberg, T. Increased prefrontal and parietal activity after training of working memory. Nat. Neurosci. 7, 75-79 (2004). • Page, T.L., Einstein, M., Duan, H., He, Y., Flores, T., Rolshud, D., Erwin, J.M., Wearne, S.L., Morrison, J.H. & Hof, P.R. Morphological alterations in neurons forming corticocortical projections in the neocórtex of aged Patas monkeys. Neuroscience Letters. 317, 37-41 (2002). • Paulo, A.C., Sampaio, A., Santos, N.C., Costa, P.S., Cunha, P., Zihl, J., Cerqueira, J., Palha, J.A. & Sousa, N. Patterns of cognitive performance in healthy ageing in northern Portugal: a cross-sectional analysis. PLoS One. 6, 24553; 10.1371/journal.pone.0024553 (2011). • Peters, R. Ageing and the brain. Postgrad. Med. J. 82, 8488 (2006). • Puram, S.V., Kim, A.H., Ikeuchi, Y., Wilson-Grady, J.T., Merdes, A., Gygi, S.P. & Bonni, A. CaMKIIβ signaling pathway at the centrosome regulates dendrite patterning in the brain. Nat. Neurosci. 14, 973-985 (2011). • Pyapali, G.K. & Turner, D.A. Increased dendritic extent in hippocampal CA1 neurons from aged F344 rats. Neurobiol. Aging. 17, 601-611 (1996). • Querfurth, H.W. & LaFerla, F.M. Alzheimer’s Disease. N. Engl. J. Med. 362, 329-344 (2010). • Ragozzino, M.E., Detrick, S. & Kesner, R.P. Involvement of the prelimbic-infralimbic areas of the rodent prefrontal cortex in behavioral flexibility for place and response learning. J. Neurosci. 19, 4585-4594 (1999). • Rampon, C., Jiang, C.H., Dong, H., Tang, Y.P., Lockhart, D.J. Schultz, P.G., Tsien, J.Z. & Hu, Y. Effects of environmental enrichment on gene expression in the brain. Proc. Natl. Acad. Sci. USA. 97, 12880-12884 (2000). • Rapp, P.R., Kansky, M.T. & Roberts, J.A. Impaired spatial information processing in aged monkeys with preserved recognition memory. Neuroreport. 8, 1923-1928 (1997). • Rapp, P.R. & Gallagher, M. Preserved neuron number in the hippocampus of aged rats with spatial learning deficits. Proc. Natl. Acad. Sci. USA. 93, 9926-9930 (1996). • Rapp, P.R., Stack, E.C. & Gallagher, M. Morphometric studies of the aged hippocampus: I. Volumetric analysis in behaviorally characterized rats. J. Comp. Neurol. 403, 459-470 (1999). • Rasmussen, T., Schliemann, T., Sørensen, J.C., Zimmer, J. & West, M.J. Memory impaired aged rats: no loss of principal hippocampal and subicular neurons. Neurobiol. Aging. 17, 143-147 (1996). • Raz, N. Neuroanatomy of aging brain: evidence from structural MRI in Neuroimaging II. Clinical Applications (ed. Bigler, E.D.) 153-182 (Academic Press,1996).
38 • Raz, N., Lindenberger, U., Rodrigue, K.M., Kennedy, K.M., Head, D., Williamson, A., Dahle, C., Gerstorf, D. & Acker, J.D. Regional brain changes in aging healthy adults: General trends, individual differences and modifiers. Cereb. Cortex. 15, 1676-1689 (2005). • Raz, N., Rodrigue, K.M. and Acker, J.D. Hypertension and the brain: vulnerability of the prefrontal regions and executive functions. Behav. Neurosci. 117, 1169-1180 (2003). • Raz, N. Aging of the brain and its impact on cognitive performance: Integration of structural and functional findings in The Handbook of Aging and Cognition (ed. Craik, F.I.M. & Salthouse, T.A.) 1-90 (Lawrence Erlbaum, 2000) • Raz, N., Ghisletta, P., Rodrigue, K. M., Kennedy, K. M. & Lindenberger, U. Trajectories of brain aging in middle-aged and older adults: regional and individual differences. Neuroimage. 51, 501-511 (2010). • Rebok, G.W., Ball, K., Guey, L.T., Jones, R.N., Kim, H.Y., King, J.W., Marsiske, M., Morris, J.N., Tennstedt, S.L., Unverzagt, F.W., Willis, S.L.; ACTIVE Study Group. Ten-year effects of the advanced cognitive training for independent and vital elderly cognitive training trial on cognition and everyday functioning in older adults. J. Am. Geriatr. Soc. 62, 16-24 (2014). • Rendeiro, C., Spencer, J.P.E., Vauzour, D., Butler, L.T., Ellis, J.A. & Williams, C.M. The impact of flavonoids on spatial memory in rodents: from behaviour to underlying hippocampal mechanisms. Genes Nutr. 4, 251-270 (2009). • Revlin, R. What is Cognitive Psychology? in Cognition: Theory and Practice (ed. Revlin, R.) 1-24 (Worth Publishers, 2012). • Richards, M., Hardy, R. & Wadsworth, M.E. Long-term effects of breast-feeding in a national birth cohort: educational attainment and midlife cognitive function. Public Health Nutr. 5, 631-635 (2002). • Robert, V., Cassim, S., Chevaleyre, V. & Piskorowski, R.A. Hippocampal area CA2: properties and contribution to hippocampal function. Cell Tissue Res. 373, 525-540 (2018). • Rubinsztein, D. C., Mariño, G. & Kroemer, G. Autophagy and aging. Cell. 146, 682-695 (2011). • Santos, N.C., Costa, P.S., Cunha, P., Cotter, J., Sampaio, A., Zihl, J., Almeida, O.F.X., Cerqueira, J.J., Palha, J.A. & Sousa, N. Mood is a key determinant of cognitive performance in communitydwelling older adults: a cross-sectional analysis. Age. 35, 1983-1993 (2013). • Satz, P. Brain reserve capacity on symptom onset after brain injury: a formulation and review of evidence for threshold theory. Neuropsychology. 7, 273-295 (1993). • Scheibel, A.B. The hippocampus: organizational patterns in health and senescence. Mech. Ageing Dev. 9, 89-102 (1979).
39 • Scheibel, M.E., Lindsay, R.D., Tomiyasu, U. & Scheibel, A.B. Progressive dendritic changes in aging human cortex. Exp. Neurol. 47, 392-403 (1975). • Scheibel, M.E., Lindsay, R.D., Tomiyasu, U. & Scheibel, A.B. Progressive dendritic changes in the aging human limbic system. Exp. Neurol. 53, 420-430 (1976). • Scheibel, M.E., Tomiyasu, U. & Scheibel, A.B. The aging human betz cell. Exp. Neurol. 56, 598-609 (1977). • Schrijver, N.C.A., Bahr, N.I., Weiss, I.C. & Wurbel, H. Dissociable effects of isolation rearing and environmental enrichment on exploration, spatial learning and HPA activity in adult rats. Pharmacol. Biochem. Behav. 73, 209-224 (2002). • Seeman, T.E., McEwen, B.S., Rowe, J.W. & Singer, B.H. Allostatic load as a marker of cumulative biological risk: MacArthur studies of successful aging. Proc. Natl. Acad. Sci. USA. 98, 4770-4775 (2001). • Seider, T.R., Fieo, R.A., O’Shea, A., Porges, E.C., Woods, A.J. & Cohen, R.A. Cognitively engaging activity is associated with greater cortical and subcortical volumes. Front. Aging Neurosci. 8, 1-10 (2016). • Shehata, M. & Inokuchi, K. Does autophagy work in synaptic plasticity and memory? Rev. Neurosci. 25, 543-557 (2014). • Shettleworth, S.J. Cognition and the study of behavior in Cognition, Evolution and Behavior (ed. Shettleworth, S.J.) 5 (Oxford University Press, 1998). • Shimada, A., Tsuzuki, M., Kelno, H., Satoh, M., Chiba, Y., Saitoh, Y. & Hosokawa, M. Apical vulnerability to dendritic retraction in prefrontal neurones of aging SAMP10 mouse: a model of cerebral degeneration. Neuropathol. Appl. Neurobiol. 32, 1-14 (2006). • Sousa, N., Cerqueira, J.J. & Almeida, O.F. Corticosteroid receptors and neuroplasticity. Brain Res. Rev. 57, 561-570 (2008). • Squire, L.R. & Wixted, J.T. The cognitive neuroscience of human memory since H.M. Annu. Rev. Neurosci. 34, 259-288 (2011). • Squire, L.R. & Zola, S.M. Structure and function of declarative and nondeclarative memory systems. Proc. Natl. Acad. Sci. USA. 93, 13515-13522 (1996). • Squire, L.R. Mechanisms of memory. Science. 232, 1612-1619 (1986). • Squire, L.R. Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol. Rev. 99, 195-231 (1992).
40 • Stein, L.R., O’Dell, K.A., Funatsu, M., Zorumski, C.F. & Izumi, Y. Short-term environmental enrichment enhances synaptic plasticity in hippocampal slices from aged rats. Neuroscience. 329, 294-305 (2016). • Stern, Y., Alexander, G.E., Prohovnik, I. & Mayeux, R. Inverse relationship between education and parietotemporal perfusion deficit in Alzheimer’s disease. Ann. Neurol. 32, 371-375 (1992). • Stern, Y. Cognitive reserve in ageing and Alzheimer's disease. Lancet Neurol. 11, 1006-1012 (2012). • Stern, Y. What is cognitive reserve? Theory and research application of the reserve concept. J. Int. Neuropsychol. Soc. 8, 448-460 (2002). • Stine-Morrow, E.A.L., Parisi, J.M., Morrow, D.G. & Park, D.C. The effects of engaged lifestyles on cognitive vitality: a field experiment. Psychol. Aging. 23, 778-786 (2008) • Strange, B.A., Witter, M.P., Lein, E.S. & Moser E.I. Functional organization of the hippocampal longitudinal axis. Nat. Rev. Neurosci. 15, 655-669 (2014). • Sugie, A., Marchetti, G. & Tavosanis, G. Structural aspects of plasticity in the nervous system of Drosophila. Neural Dev. 13, 14; 10.1186/s13064-018-0111-z (2018). • Sullivan, E.V., Marsh, L., Mathalon, D.H., Lim, K.O. & Pfefferbaum, A. Age-related decline in MRI volumes but of temporal lobe gray matter but not hippocampus. Neurobiol. Aging. 16, 591-606 (1995). • Syková, E., Mazel, T., Hasenöhrl, R.U., Harvey, A.R., Simonová, Z., Mulders, W.H. & Huston, J.P. Learning deficits in aged rats related to decrease in extracellular volume and loss of diffusion anisotropy in hippocampus. Hippocampus. 12, 269-279 (2002). • Tammes, C.K., Walhovd, K.B., Dale, A.M., Ostby, Y., Grydeland, H., Richardson, G., Westlye, L.T., Roddey, J.C., Hagler, D.J., Jr., Due-Tonnessen, P., Holland, D., Fjell, A.M. & Alzeimer´s Disease Neuroimaging, I. Brain development and aging: overlapping and unique patterns of change. Neuroimage. 68, 63-74 (2013). • Tang, G., Gudsnuk, K., Kuo, S.H., Cotrina, M.L., Rosoklija, G., Sosunov, A., Sonders, M.S., Kanter, E., Castagna, C., Yamamoto, A., Yue, Z., Arancio, O., Peterson, B.S., Champagne, F., Dwork, A.J., Goldman, J. & Sulzer, D. Loss of mTOR-dependent macroautophagy causes autistic-like synaptic pruning deficits. Neuron. 83, 1131-1143 (2014). • Tanigawa, T., Takechi, H., Arai, H., Yamada, M., Nishiguchi, S. & Aoyama, T. Effect of physical activity on memory function in older adults with mild Alzheimer’s disease and mild cognitive impairment. Geriatr. Gerontol. Int. 14, 758-762 (2014).
41 • Tank, E.M.H., Rodgers, K.E. & Kenyon, C. Spontaneous Age-related neurite branching in caenorhabditis elegans. J. Neurosci. 31, 9279-9288 (2011). • Teffer, K. & Semendeferi, K. Human prefrontal cortex: evolution, development, and pathology. Prog. Brain Res. 195, 191-218 (2012). • Tierney, P.L., Degenetais, E., Thierry, A.M., Glowinski, J. & Gioanni, Y. Influence of the hippocampus on interneurons of the rat prefrontal cortex. Eur. J. Neurosci. 20, 514-524 (2004). • Tosato, M., Zamboni, V., Ferrini, A. & Cesari, M. The aging process and potential interventions to extend life expectancy. Clin. Interv. Aging. 2, 401-412 (2007). • Toth, M., Melentijevic, I., Shah, L., Bhatia, A., Lu, K., Talwar, A., Naji, H., Ibanez-Ventoso, C., Ghose, P., Jevince, A., Xue, J., Herndon, L.A., Bhanot, G., Rongo, C., Hall, D.H. & Driscoll, M. Neurite sprouting and synapse deterioration in the aging c.elegans nervous system J. Neurosci. 27, 87788790 (2012). • Turner, D.A. & Deupree, D.L. Functional elongation of CA1 hippocampal neurons with aging in Fischer 344 rats. Neurobiol. Aging. 12, 201-210 (1991). • Uttl, B. & Graf, P. Episodic spatial memory in adulthood. Psychol. Aging. 8, 257-273 (1993). • Uylings, H.B.M. & de Brabander, J.M. Neuronal changes in normal human aging and Alzheimer´s disease. Brain Cogn. 49, 268-276 (2002). • Uylings, H.B., Groenewegen, H.J. & Kolb, B. Do rats have a prefrontal cortex?. Behav. Brain Res. 146, 3-17 (2003). • Van der Staay, F.J. Spatial working memory and reference memory of Brown Norway and WAG rats in a holeboard discrimination task. Neurobiol. Learn. Mem. 71, 113-125 (1999). • Van der Staay, F.J., Gieling, E.T., Pinzón, N.E., Nordquist, R.E. & Ohl, F. The appetitively motivated "cognitive" holeboard: a family of complex spatial discrimination tasks for assessing learning and memory. Neurosci Biobehav. Rev. 36, 379-403 (2012). • Van Eden, C.G. & Uylings, H.B.M. Cytoarchitectonic development of the prefrontal cortex in the rat. J. Comp. Neurol. 241, 253-267 (1985). • Van Petten, C. Relationship between hippocampal volume and memory ability in healthy individuals across the lifespan: review and meta-analysis. Neuropsychologia . 42, 1394-1413 (2004). • Vertes, R.P., Hoover, W.B., Szigeti-Buck, K. & Leranth, C. Nucleus reuniens of the midline thalamus: link between the medial prefrontal cortex and the hippocampus. Brain Res. Bull. 71, 601-609 (2007).
42 • Vicens, P., Bernal, M.C., Carrasco, M.C. & Redolat, R. Previous training in the water maze: differential effects in NMRI and C57BL mice. Physiol. Beh. 67, 197-203 (1999). • Vicens, P., Redolat, R. & Carrasco, M.C. Effects of early spatial training on water maze performance: a longitudinal study of mice. Exp. Gerontol . 37, 575-581 (2002). • Wang, G.W. & Cai, J.X. Disconnection of the hippocampal-prefrontal cortical circuits impairs spatial working memory performance in rats. Behav. Brain Res. 175, 329-336 (2006). • Wang, G.W. & Cai, J.X. Reversible disconnection of the hippocampal-prelimbic cortical circuit impairs spatial learning but not passive avoidance learning in rats. Neurobiol. Learn. Mem. 90, 365-373 (2008). • Whalley, L.J., Deary, I.J., Appleton, C.L. & Starr, J.M. Cognitive reserve and the neurobiology of cognitive aging. Ageing Res. Rev. 3, 369-382 (2004). • Willis, S.L., Tennstedt, S.L., Marsiske, M., Ball, K., Elias, J., Koepke, K.M., Morris, J.N., Rebok, G.W., Unverzagt, F.W., Stoddard, A.M., Wright, E., ACTIVE Study Group. Long-term effects of cognitive training on everyday functional outcomes in older adults. JAMA. 296, 2805-2814 (2006). • Wilson, R.S., Bennett, D.A., Bienias, J.L., Aggarwal, N.T., Mendes de Leon, C.F., Morris, M.C., Schneider, J.A. & Evans, D.A. Cognitive activity and incident AD in a population based sample of older persons. Neurology. 59, 1910-1914 (2002). • Wirt, R.A. & Hyman, J.M. Integrating spatial working memory and remote memory: interactions between the medial prefrontal cortex and hippocampus. Brain Sci. 7, 43; 10.3390/brainsci7040043 (2017). • Wong, R.O.L. & Ghosh, A. Activity-dependent regulation of dendritic growth and patterning. Nat. Rev. Neurosci. 3, 803-812 (2002). • Yankner, B.A., Lu, T. & Loerch, P. The aging brain. Annu. Rev. Pathol. 3, 41-66 (2008). • Yates, M.A., Markham, J.A., Anderson, S.E., Morris, J.R. & Juraska, J.M. Regional variability in agerelated loss of neurons from the primary visual cortex and medial prefrontal cortex of male and female rats. Brain Res. 1218, 1-12 (2008). • Yin, J. & Yuan, Q. Structural homeostasis in the nervous system: a balancing act for wiring plasticity and stability. Front. Cell Neurosci. 8, 439; 10.3389/fncel.2014.00439 (2015). • Zilles, K. & Wree, A. Cortex: areal and laminar structure in The Rat Nervous System , (ed. Paxinos, G.) 729-757 (Academic Press, 2004).
43 Chapter II Structural and molecular correlates of cognitive aging in the rat Mota C, Taipa R, Pereira das Neves S, Monteiro-Martins S, Monteiro S, Palha JA, Sousa N, Sousa JC & Cerqueira JJ Manuscript published in Scientific Reports in February 2019
50 Figure 2| Morphological analysis of HPC neuron dendritic arborizations. When a random sample of all animals is considered (older = 27; younger = 15): A, B and C) Comparison of dendritic lengths of DG granular, CA3 and CA1 pyramidal neurons between younger and older rats. D) Correlation between granular neuron dendritic lengths and individual performances in the reference memory task of both younger and older rats. When similar age animals are clustered (see methods for details) in Good and Bad performers according to reference memory performance (older GP = 16 (18 for CA1); older BP = 8 (9 for CA1); younger GP = 10; younger BP = 5): E) Average dendritic lengths for both GPs and BPs of younger and older animals. F) Sholl analysis of the apical dendrite of DG granular neurons. This graph presents the mean number of intersections of apical dendritic branches with consecutive 20µm spaced concentric spheres. G) Representative reconstructions of DG granular neurons used in the previous analysis. H, I, J and K) The same analysis was performed for CA3 pyramidal neurons and in L, M, N
51 and O for CA1 pyramidal neurons. Error bars represent SEM, doted lines represent confidence intervals and continuous lines are linear fits; * p <0.05; ** p <0.01; *** p <0.001. (RM – reference memory) Regarding comparisons within age groups, aged BPs presented a significant increase in the dendritic length of both granular and apical dendrite of CA1 pyramidal neurons when compared with aged GPs (DG: t (22)=-2.632, p =0.015, d =1.033; CA1 apical dendrite: t (25)=-3.312, p =0.003, d =1.382) (Fig. 2e,m). Also, regarding granular neurons, a significant difference was observed between young GPs and BPs. Here, GPs display higher dendritic lengths when compared with BPs ( t (13)=3.540, p =0.004, d =1.952) (Fig. 2e). Data on CA3 pyramidal neurons revealed a significant effect of age, but not of performance group nor any interaction, in the length of both basal and apical dendrites (Table1, Fig. 2i). To explore in which parts of the dendritic tree laid the above-mentioned differences, we performed a Sholl analysis, which measures the number of intersections as a function of distance from the soma. Results for granular dendrites revealed a significant effect of age, but not of performance group, and a significant interaction between the two (Table 1, Fig. 2f). Repeated measures ANOVA revealed that younger GPs, when compared with the BP group, had an overall increase in the number of intersections ( F (1,13)=11.050, p =0.005, ηρ²=0.459), both proximally and distally (Fig. 2f). Results of the Two-way ANOVA analysis for CA3 and CA1 apical dendrites revealed no significant effect of age, performance, neither an interaction between these two factors (Fig. 2j,n). However, group comparisons revealed an overall increase in the number of intersections in apical CA1 dendrites of young GPs when compared with the BP group ( F (1,13)=6.294, p =0.026, ηρ²=0.326) (Fig. 2n). These alterations observed in HPC neurons are exemplified in the reconstructions of figures 2g,k,o. Since some of the cognitive tasks assessed in this work were mPFC-dependent, we also analyzed the morphology of mPFC neurons (Cg/PL and IL pyramidal neurons; Supplementary Fig. S4). We found an age-dependent reduction in the length of basal dendrites of Cg/PL pyramidal neurons, but no major changes in other dendritic domains (Supplementary Fig. S4a,b). Pearson correlations between behavioral performance and dendritic length showed a significant association between working memory and the apical dendritic length of Cg/PL, that was positive in younger subjects (r2 = 0.630, p = 0.012) and negative in older animals (r2 = -0.504, p = 0.007) (Supplementary Fig. S4c). The performance in the reference memory task was only significantly negatively correlated with the apical dendritic length of IL pyramidal neurons of older animals (r2 = -0.465, p = 0.029) (Supplementary Fig. S4o). Regarding the behavioral flexibility task, only in younger animals a positive correlation was found between this task and the apical dendritic length of IL pyramidal neurons (r=0.611, p=0.046; Supplementary Table S5; for additional
52 information regarding individual animal performance in all cognitive tasks see Supplementary Fig. S3e,f). In addition to individual correlations, we performed within group comparisons (older and younger GPs and BPs for each task) on average dendritic lengths and distribution of dendritic processes. Interestingly, differences were only present between IL neurons of GPs and BPs in younger animals ( t (10)=2.377, p =0.039, d =1.550) (Supplementary Fig. S4p). Finally, the distribution of dendritic processes resulting from Sholl analysis in mPFC neurons showed a significantly more ramified apical dendritic tree of IL pyramidal neurons in younger GPs as compared with younger BPs ( F (1,10)=9.339, p =0.012, ηρ²=0.483), with no differences in the other parameters (Supplementary Fig. S4q). For a comprehensive overview of the simultaneous alterations occurring at different brain regions, Supplementary Fig. S6 depicts the individual morphological alterations of the analyzed animals, including animal performance and respective relative dendritic lengths of DG, CA3, CA1, Cg-PL and IL brain regions. 3.3. Impaired autophagy impacts on dendritic structure Autophagic activity has been identified as a critical mechanism underlying dendritic remodeling25. To test the hypothesis that autophagy signaling is disrupted in aging and could be associated with alterations in dendritic recycling in the HPC, we performed western blot analysis of the autophagosome markers LC3 and p62. To determine the relationship between autophagic activity and dendritic size, protein levels of BDNF were analyzed. As for structural markers of synaptic function the levels of PSD95, SNAP25 and SYP were determined (Fig. 3c). To test the association between dendritic length and autophagy/neurotrophin levels, the levels of the synaptic marker PSD95 (a surrogate marker of dendritic extension) was correlated with both p62 (whose increased levels represent decreased autophagic activity) and BDNF. In both younger and older animals, HPC levels of PSD95 were positively correlated with p62 (younger: r =0.850, p =0.001; older r =0.878, p <0.0005; Fig. 3j) and BDNF (younger: r =0.671, p =0.024; older r =0.692, p =0.001; Fig. 3k).
53 Figure 3| Defective autophagy signaling and dendritic pruning in the HPC of older BPs. Performance in reference memory was used to cluster (see methods for details) both younger and older animals as GPs and BPs. A random sample of these were used for molecular analyses (younger GPs = 5-6; younger BPs = 5; older GPs = 10; older BPs = 9-10). A and B) Levels of autophagy markers, LC3-II (A) and p62 (B), normalized to actin. D) BDNF levels normalized to tubulin. E, F, G) Levels of synaptic markers PSD95, SYP, and SNAP25 normalized to actin, tubulin, and tubulin, respectively. C) Representative western blots of PSD95, p62, SNAP25, LC3, Actin, Tubulin, SYP, and BDNF. For each protein, the blots were cropped from different parts of the same gel. H and I) Correlation between RM performance and p62 or BDNF levels, respectively. J and K) Correlation between PSD95 and p62 or BDNF levels, suggesting a relationship between the levels of synaptic markers and autophagy or dendritic growth, respectively. Error bars represent SEM, doted lines represent confidence intervals and continuous lines are linear fits; * p <0.05. (RM – reference memory)
54 Two-way ANOVA revealed a significant effect of reference memory performance, but not of age, neither an interaction between them, in the HPC levels of LC3-II, but not of p62 (Fig. 3a,b; Table 1). Group comparisons further revealed that, in younger animals, the levels of LC3-II in the HPC were significantly lower in BPs than in GPs ( t (9)=2.988, p =0.015, d =1.750; Fig. 3a), while the levels of p62 in the HPC were similar between the two groups. In older animals there was a trend toward BPs animals having less HPC LC3-II ( t (18)=2.020, p =0.059, d =0.903); Fig. 3a) and more p62 ( t (18)=-1.912, p =0.072, d =0.855; Fig. 3b). At the individual level, younger animals had a significant positive correlation between performance in the reference memory task and the level of HPC LC3-II ( r =0.790, p =0.004), but not of p62, while in older rats there was a significant correlation between reference memory performance and the levels of both HPC LC3-II ( r =0.523, p =0.018) and p62 ( r =-0.489, p =0.029; Fig. 3h) (see also Supplementary Table S7). There were no significant correlations between any HPC autophagy marker and performance in the working memory task or behavioral flexibility tasks (Supplementary Table S7). Hippocampal BDNF protein levels were similar in younger BPs and GPs groups, but were significantly higher in older BPs than in older GPs ( t (18)=-2.500, p =0.022, d =1.118; Fig. 3d). In younger animals, there was also a trend toward a positive correlation between HPC BDNF levels and performance in working memory task ( r =0.610, p =0.061) but not in the reference memory task ( r =0.382, p =0.247; Fig. 3i) nor in the behavioral flexibility task (Supplementary Table S7). In older rats, HPC BDNF levels were significantly negatively correlated with the performance in reference and working memory tasks (reference memory: r =-0.577, p =0.008; Fig. 3i; working memory: r =-0.447, p =0.048) but not the behavioral flexibility task (Supplementary Table S7). Regarding synaptic markers, only HPC SNAP25 levels (Fig. 3g, Table 1) had a significant interaction between age and performance group, without significant effect of either factor alone. In addition, HPC levels of PSD95 (Fig. 3e), SYP (Fig. 3f) and SNAP25 (Fig. 3g) were different between GPs and BPs groups solely in older animals. In line with data from BDNF, a significant increase was observed in the levels of SNAP25 and a trend towards an increase of PSD95 and SYP in older BPs, when compared with older GPs (SNAP25 t (18)=-2.192, p =0.042, d =0.980; PSD95 t (18)=-2.045, p =0.056, d =0.914; SYP t (17)=- 2.042, p =0.057, d =0.912). Lastly, concerning older animals, HPC PSD95 levels showed a significant negative correlation with reference ( r =-0.448, p =0.048) and working memory ( r =-0.513, p =0.021) performances, while HPC SNAP25 levels presented solely a significant negative correlation with the performance in the reference memory task (reference memory task: r =-0.560, p =0.010, working memory task: r =-0.339, p =0.143). For HPC SYP levels, a trend toward a negative correlation with reference memory ( r =-0.453, p =0.052) and working memory tests ( r =-0.406, p =0.085) was observed. No
55 correlations were found between synaptic markers and the performance in the behavioral flexibility task (Supplementary Table S7; for additional information regarding individual animal performance in all cognitive tasks see Supplementary Fig. S3d). As previously described for the HPC, we performed an analysis of autophagicand dendritic growth-related proteins in the mPFC (Supplementary Fig. S8). When animals were grouped by performance in the working memory, within group analysis revealed that in younger animals, levels of LC3-II and p62 were similar between GPs and BPs whereas in older BPs, there was a significant decrease in the levels of LC3II ( t (16)=2.197, p =0.043, d =1.045) and a significant increase in the levels of p62 ( t (17)=-3.326, p =0.004, d =1.515), suggesting a lower level of autophagy in older BPs (Supplementary Fig. S8a,b). Also, older animals, but not younger animals, presented a negative correlation between p62 and the performance in reference ( r =-0.504, p =0.028) and working memory task ( r =-0.646, p =0.003; Supplementary Fig. S8h), but not the behavioral flexibility task (Supplementary Table S9). Regarding neurotrophins and synaptic markers, mPFC BDNF, PSD95 and SYP protein levels were significantly higher in older BPs compared with older GPs (BDNF t (18)=-2.226, p =0.039, d =0.995; PSD95 t (16)=-1.959, p =0.068, d =0.937; SYP t (18)=-2.108, p =0.049, d =0.943) (Supplementary Fig. S8d,e,f). Also, BDNF levels were not correlated with performance in the working memory (Supplementary Fig. S8i), reference memory or behavioral flexibility (Supplementary Table S9) in any age group, and were correlated with PSD95 in older animals ( r = 0.671, p =0.002; Supplementary Fig. S8k) but not younger rats ( r =-0.256, p =0.476; Supplementary Fig. S8k; for additional information regarding individual animal performance in all cognitive tasks see Supplementary Fig. S3g). This data is in agreement to what we previously described for the HPC, further showing that the levels of autophagy in mPFC neurons are strongly and positively related with dendritic pruning and better performances in older individuals. 4. Discussion The present study addressed the heterogeneity of cognitive-aging from a multidimensional perspective to reveal a hitherto of unappreciated complexity. It explored its underpinnings, highlighting, for the first time, the role of the balance between neurotrophic and autophagic activities in such processes. Data herein presented confirmed that, despite a general aging-associated cognitive decline, the performance of both young adult and older rats in working and reference memory tasks is heterogeneous, particularly in older individuals2,35; in the latter, we confirmed that a certain proportion of subjects maintain spatial memory abilities comparable to those of younger animals6,7. Of notice, the data clearly revealed, for the first time,
56 that this age-associated increase in the dispersion of individual performance is not universal, as it was not present in all cognitive dimensions (e.g. the behavioral flexibility task). The working and reference memory tasks both assess spatial learning and memory, albeit within a different timeframe: while the former is dependent on short-term memory and HPC to mPFC connections29,36, the latter depends on long-term memory and largely on the integrity of the HPC28. In contrast, the behavioral flexibility task is memory independent and assesses the ability to adapt to changing circumstances37, a critical component of executive function, which is a very distinct cognitive ability. In light of this distinction, the observation that aging is accompanied by an increased interindividual heterogeneity (and a performance decline, on average) in memory-dependent but not executivefunction-dependent tasks adds yet another layer of heterogeneity to the aging process, and strongly suggests that some cognitive functions (and the networks sub-serving them) are more prone to aging than others. Significantly, this seems also to be the case in humans38 in which both long-term and working memory are more influenced by age-related impairments than knowledge of vocabulary and priming, a form of non-declarative memory. Despite this, the herein reported relative preservation of executive function in rodents might seem to contradict several studies in humans showing that executive processes are also disrupted in aging39-41. However, it is important to highlight that, besides the obvious species difference, executive function tasks in humans are often contaminated by deficits in speed of processing42 which are well known to be affected by aging. On the contrary, in our experiments, not only was average swimming speed similar in all older subjects (and not significantly different from that of their younger counterparts) but also the behavioral flexibility test is independent of it. In addition to the two layers of heterogeneity in aging discussed above, analysis of each animal’s membership to either the GPs or BPs group, for each behavioral test, further revealed another dimension of inter-individual heterogeneity. Indeed, when comparing cluster membership for each individual in each test, we showed for the first time that most animals were GPs in some tests and BPs in others, without a clear separating pattern in either younger or older groups. Moreover, the overall distribution of animals according to their group membership (GPs or BPs) for the 3 tests (working and reference memory and behavioral flexibility) was strikingly similar in both age groups, with only 13% of older and 19% of younger animals being good in every task and 22% of older and 19% of younger animals being bad in every task. More importantly, this third heterogeneity level seems to be independent of the other two. In other words, despite all the inter-individual heterogeneity (within a given test) and the heterogeneity between tests (with reference and working memory being more sensitive to aging than behavioral flexibility) there is also heterogeneity, at the individual level, as to being a GP or BP for each behavioral test.
57 In summary, the behavioral data suggest that cognitive decline in aging is not inevitable, or strictly linked to chronological age and that, even in a relatively homogeneous population of animals such as the one in this study, there is a high variability and complexity in the way the different cognitive functions are preserved/impaired in each individual. Understanding the underpinnings of individual differences may help to explain the observed heterogeneity and, possibly, what determines the existence of healthier agers, which was next studied at the morphological and molecular levels. Changes in the morphology of neuronal dendritic trees were shown to correlate with cognitive alterations in both rodents and humans43,44. In aging, it is already well established that memory impairments are not related with neuronal loss35,45 but rather to volume changes17 and altered morphology of neuronal dendritic trees46,47. In the present work, using a large number of rats (15 young and 27 old) we showed that, on average, older animals have shorter apical dendritic arborizations in dorsal HPC neurons (dentate gyrus granules and CA1 and CA3 pyramids) but similar apical dendritic trees in mPFC neurons (Cg/PL and IL layers II/III pyramids) when compared to younger animals. These findings are in line with most previous studies that analyzed one or the other region (HPC48-53; mPFC54-56) and might suggest that the frontal regions might be less affected by the aging process or that age-related changes in neuronal morphology appear later in the mPFC. This is partly corroborated by the fact that age-related apical dendritic retraction in the mPFC was only reported in one study57. Interestingly, this relative mPFC "resilience" might be specific for the superficial layers, since deeper, layer V, pyramidal neurons, similar to hippocampal cells, exhibit agerelated apical dendritic retraction at 20-22 months20,53,56. Importantly, this has been observed in humans, in which age-related dendritic retraction, in the same individuals, was 3 times more prominent in the deep than in the superficial PFC pyramids58. The fact that mPFC dendrites are less affected by aging fits perfectly with the behavioral data presented here, pointing to an attenuated age-associated decline of executive functions. Another major novelty of the present work is the finding that older animals with deficits in HPC-dependent tasks have larger dendritic trees in the HPC than cognitively intact rats of the same age. Some previous papers had already shown that hippocampal cells from older rats59 and older humans60,61 had increased dendritic length, but in none of these studies were subjects cognitively characterized. Interestingly, in the mPFC, despite no overall age-related retraction, there was a similar, albeit with smaller magnitude, association between bigger apical dendritic trees in layer II/III Cg/PL pyramids and worse performance in the working memory (mPFC-dependent) test. This association, in older animals, between larger dendritic trees and poorer cognitive function might be considered contra-intuitive. However, while bigger dendritic trees might mean more connectivity and better neuronal function, it is also well described that
58 the accumulation of "waste" dendritic material and large dendritic trees, for example as a result of impaired autophagy and dendritic pruning deficits, hampers neuronal function and correlates with cognitive deficits, in both humans and animals23. Interestingly, in Fragile-X-syndrome patients, who have impaired dendritic pruning, there is also an inverse correlation between hippocampal volume and cognitive performance, which is not present in age-matched individuals without pruning deficits62. Of note, in the present work, the inverse correlations between dendritic tree length and cognitive performance are not present in the group of younger adults (in which an opposite trend is observed), supporting an agerelated phenomenon. Together, these results suggest that the inverse correlation between large dendritic trees and poor cognitive performance in the elderly, might be attributed to age-associated dendritic pruning deficits leading to larger, less efficient, dendritic trees. In light of this hypothesis, individual differences in dendritic pruning might also underpin the individual heterogeneity in cognitive aging. Dendritic pruning is a mechanism often used to selectively remove unnecessary and exuberant neuronal branches, not only in the immature nervous system23 but also in the adult HPC63, thus ensuring the proper formation of functional optimized circuitries. Dendritic and synaptic pruning is highly dependent on autophagy-dependent protein turnover, as animals presenting constitutional23 or induced64 inhibition of autophagy have larger dendritic trees and increased spine density, which correlate with cognitive deficits. In order to further dissect whether this could contribute to the observed morphology, we analyzed the levels of autophagic activity in the HPC and mPFC. Furthermore, this was complemented with a quantification of the neurotrophin BDNF, a main inducer of dendritic and spine growth65. Finally, given the technical challenge to assess protein levels and dendritic tree length in the same region of the same animals, these results were correlated with preand post-synaptic markers. Indeed, there is a consensus in the literature that these levels, particularly when concordant, are a good surrogate of synaptic abundance and dendritic tree complexity66-68. In support of this assumption, here we show that older, but not young, cognitively impaired animals have higher levels of these synaptic proteins than age-matched cognitively intact rats, precisely replicating the findings from the dendritic tree analysis. With the present work, we reveal that older cognitively impaired animals have reduced autophagic activity in both the dorsal HPC and the mPFC, when compared with older cognitively intact rats. Of notice is the fact that a decrease in the relative abundance of the autophagic vacuole marker LC3-II (lipidated LC3) was accompanied by a correspondent increase in the relative abundance of the autophagy cargo-protein p6269, attesting the robustness of the findings. Significantly, these observations were specific for older animals, as levels of autophagic markers in both brain regions did not differ between cognitively intact and cognitively impaired younger adult individuals. In most organisms, pathological aging is associated
59 with decreased autophagic activity and autophagy inhibition induces degenerative changes that resemble those associated with aging70. While the mechanisms of such relationship are far from being well understood, the most prevalent hypothesis considers a failure to clean toxic/waste protein debris, that accumulate with time and induce cellular dysfunction70. In line with this, we found that, in the older, decreased levels of autophagy are strongly and inversely correlated with the abundance of synaptic markers, a surrogate marker of dendritic length. Our findings, however, further extend the interpretation of the previous observations, by suggesting that in neurons, decreased autophagy results in less dendritic pruning and an accumulation of dendrites that hamper neuronal function. Significantly, this does not seem to be an inevitable consequence of aging, as levels of autophagic and synaptic markers in older cognitively intact individuals were strikingly similar to those of younger animals. Of note, BDNF levels similarly did not vary significantly with aging but were also increased in older cognitively impaired, compared with cognitively intact animals, suggesting that increased dendritic growth, as well as decreased autophagic activity, might also contribute to the increased dendritic length observed in these animals. Many factors could induce a decreased autophagic activity, similar to that presented by older cognitively impaired individuals. One of the best candidates is an enhanced activity of the mechanistic target of rapamycin (mTOR, formerly known as mammalian target of rapamycin mTOR) complex, a redox/energy/nutrient sensor that inhibits autophagy and stimulates protein synthesis71. Increased mTOR activity (resulting in decreased autophagic activity) has been linked to cognitive dysfunction and learning deficits in a variety of disorders72,73, which are also associated with an increase in dendritic spines74. More importantly, and in line with our results, lifelong treatment of mice75,76 or accelerated senescence rats77 with the mTOR inhibitor rapamycin (that is considered an autophagy inducer) improved age-related cognitive dysfunction. Given the above, it is plausible to conclude that an impairment of neuronal autophagic activity could result in a scarcity of pruning mechanisms in aging neural circuits, leading to an accumulation of dendritic material and to the consequent decrease of cognitive performance. Other factors that are commonly associated with aging could also impact dendritic length, including altered glutamatergic transmission and insulin signaling. However, since these would ultimately lead to changes in neurotrophins and/or autophagic processes, we did not address these separately in the present work. Nevertheless, in order to gain full insight into the individual determinants of altered autophagic activity, these and other factors should be taken in consideration. Furthermore, insights into the relevance of all these mechanisms can only be obtained by experimental manipulations of autophagy, which were not the scope of the present work but should be pursued in the future.
66 8. Gallagher, M., Burwell, R. & Burchinal, M. Severity of spatial learning impairment in aging: development of a learning index for performance in the Morris water maze. Behav. Neurosci. 107, 618-626 (1993). 9. Aggleton, J. P., Blindt, H. S. & Candy, J. M. Working memory in aged rats. Behav. Neurosci. 103, 975-983 (1989). 10. Luine, V. & Hearns, M. Spatial memory deficits in aged rats: contributions of the cholinergic system assessed by ChAT. Brain Res. 523, 321-324 (1990). 11. Bimonte, H. A., Nelson, M. E. & Granholm, A. C. Age-related deficits as working memory load increases: relationships with growth factors. Neurobiol. Aging. 24, 37-48 (2003). 12. Barrett, G. L., Bennie, A., Trieu, J., Ping, S. & Tesafoulis, C. The chronology of age-related spatial learning impairment in two rat strains, as tested by the Barnes maze. Behav. Neurosci. 123, 533538 (2009). 13. Driscoll, I. et al. The aging hippocampus: cognitive, biochemical and structural findings. Cereb. Cortex. 13, 1344-1351 (2003). 14. Driscoll, I. et al. Longitudinal pattern of regional brain volume change differentiates normal aging from MCI. Neurology. 72, 1906-1913 (2009). 15. Freeman, S. H. et al. Preservation of neuronal number despite age-related cortical brain atrophy in elderly subjects without Alzheimer disease. J. Neuropathol. Exp. Neurol. 67, 1205-1212 (2008). 16. Raz, N., Ghisletta, P., Rodrigue, K. M., Kennedy, K. M. & Lindenberger, U. Trajectories of brain aging in middle-aged and older adults: regional and individual differences. Neuroimage. 51, 501-511 (2010). 17. Rapp, P. R., Stack, E. C. & Gallagher, M. Morphometric studies of the aged hippocampus: I. Volumetric analysis in behaviorally characterized rats. J. Comp. Neurol. 403, 459-470 (1999). 18. Driscoll, I. et al. The aging hippocampus: a multi-level analysis in the rat. Neuroscience . 139, 11731185 (2006). 19. de Brabander, J. M., Kramers, R. J. & Uylings, H. B. Layer-specific dendritic regression of pyramidal cells with ageing in the human prefrontal cortex. Eur. J. Neurosci. 10, 1261-1269 (1998). 20. Markham, J. A. & Juraska, J. M. Aging and sex influence the anatomy of the rat anterior cingulate cortex. Neurobiol. Aging. 23, 579-588 (2002). 21. Dickstein, D. L, Weaver, C. M., Luebke, J. I. & Hof, P. R. Dendritic spine changes associated with normal aging. Neuroscience. 251, 21-32 (2013). 22. Querfurth, H. W. & LaFerla, F. M. Alzheimer’s Disease. N. Engl. J. Med. 362, 329-344 (2010).
67 23. Tang, G. et al. Loss of mTOR-dependent macroautophagy causes autistic-like synaptic pruning deficits. Neuron. 83, 1131-1143 (2014). 24. Puram, S. V. et al. CaMKIIβ signaling pathway at the centrosome regulates dendrite patterning in the brain. Nat. Neurosci. 14, 973-985 (2011). 25. Bingol, B. & Sheng, M. Deconstruction for reconstruction: the role of proteolysis in neural plasticity and disease. Neuron. 69, 22-32 (2011). 26. Kanamori, T., Yoshino, J., Yasunaga, K., Dairyo, Y. & Emoto, K. Local endocytosis triggers dendritic thinning and pruning in Drosophila sensory neurons. Nat. Commun. 6, 6515 (2015). 27. Cerqueira, J. J., Mailliet, F., Almeida, O. F., Jay, T. M. & Sousa, N. The prefrontal cortex as a key target of the maladaptive response to stress. J. Neurosci. 27, 2781-2787 (2007a). 28. Morris, R. Developments of a water-maze procedure for studying spatial learning in the rat. J. Neurosci. Methods. 11, 47-60 (1984). 29. Kesner, R. P. Subregional analysis of mnemonic functions of the prefrontal cortex in the rat. Psychobiology. 28, 219-228 (2000). 30. Gibb, R. & Kolb, B. A method for vibratome sectioning of Golgi-Cox stained whole rat brain. J. Neurosci. Methods. 79, 1-4 (1998). 31. Pinto, V. et al. Differential impact of chronic stress along the hippocampal dorsal–ventral axis. Brain Struct. Funct. 220, 1205-1212 (2015). 32. Cerqueira, J. J., Taipa, R., Uylings, H. B., Almeida, O. F. & Sousa, N. Specific configuration of dendritic degeneration in pyramidal neurons of the medial prefrontal cortex induced by differing corticosteroid regimens. Cereb. Cortex. 17, 1998-2006 (2007b). 33. Uylings, H. B., Ruiz-Marcos, A. and van Pelt, J. The metric analysis of three-dimensional dendritic tree patterns: a methodological review. J. Neurosci. Methods. 18, 127-151 (1986). 34. Uylings, H. B. & van Pelt, J. Measures for quantifying dendritic arborizations. Network. 13, 397-414 (2002). 35. Rapp, P. R. & Gallagher, M. Preserved neuron number in the hippocampus of aged rats with spatial learning deficits. Proc. Natl. Acad. Sci. USA. 93, 9926-9930 (1996). 36. Goldman-Rakic, P. S. Architecture of the prefrontal cortex and the central executive. Ann. N. Y. Acad. Sci. 769, 71-83 (1995). 37. de Bruin, J. P., Sànchez-Santed, F., Heinsbroek, R. P., Donker, A. & Postmes, P. A behavioural analysis of rats with damage to the medial prefrontal cortex using the Morris water maze: evidence for behavioural flexibility, but not for impaired spatial navigation. Brain Res. 652, 323-333 (1994).
68 38. Buckner, R. L. Memory and executive function in aging and AD: multiple factors that cause decline and reserve factors that compensate. Neuron. 44, 195-208 (2004). 39. Schacter, D. L., Kaszniak, A. W., Kihlstrom, J. F. & Valdiserri, M. The relation between source memory and aging. Psychol. Aging. 6, 559-568 (1991). 40. Johnson, M. K., Hashtroudi, S. & Lindsay, D. S. Source monitoring. Psychol. Bull. 114, 3-28 (1993). 41. West, R. L. An application of prefrontal cortex function theory to cognitive aging. Psychol Bull. 120, 272-292 (1996). 42. Head, D., Kennedy, K. M., Rodrigue, K. M. & Raz, N. Age differences in perseveration: cognitive and neuroanatomical mediators of performance on the Wisconsin Card Sorting Test. Neuropsychologia. 47, 1200-1203 (2009). 43. Cerqueira, J. J. et al. Morphological correlates of corticosteroid-induced changes in prefrontal cortexdependent behaviors. J. Neurosci. 25, 7792-7800 (2005). 44. Becker, L. E, Armstrong, D. L. & Chan, F. Dendritic atrophy in children with Down's syndrome. Ann. Neurol. 20, 520-526 (1986). 45. Rasmussen, T., Schliemann, T., Sørensen, J. C., Zimmer, J.& West, M. J. Memory impaired aged rats: no loss of principal hippocampal and subicular neurons. Neurobiol. Aging. 17, 143-147 (1996). 46. Burke, S. N. & Barnes, C. A. Neural plasticity in the ageing brain. Nat. Rev. Neurosci. 7, 30-40 (2006). 47. Dickstein, D. L. et al. Changes in the structural complexity of the aged brain. Aging Cell. 6, 275-284 (2007). 48. Geinisman, Y., Bondareff, W. & Dodge, J. T. Dendritic atrophy in the dentate gyrus of the senescent rat. Am. J. Anat. 152, 321-329 (1978). 49. Machado-Salas, J. P. & Scheibel, A. B. Limbic system of the aged mouse. Exp. Neurol. 63, 347-355 (1979). 50. Lolova, I. Dendritic changes in the hippocampus of aged rats. Acta Morphol. Hung. 37, 3-10 (1989). 51. Luebke, J. I. & Rosene, D. L. Aging alters dendritic morphology, input resistance, and inhibitory signaling in dentate granule cells of the rhesus monkey. J. Comp. Neurol. 460, 573-584 (2003). 52. Markham, J. A., McKian, K. P., Stroup, T. S. & Juraska, J. M. Sexually dimorphic aging of dendritic morphology in CA1 of hippocampus. Hippocampus. 15, 97-103 (2005). 53. Chen, J. R., Tseng, G. F., Wang, Y. J. & Wang, T. J. Exogenous dehydroisoandrosterone sulfate reverses the dendritic changes of the central neurons in aging male rats. Exp. Gerontol. 57, 191202 (2014).
69 54. Anderson, R. M., Birnie, A. K., Koblesky, N. K., Romig-Martin, S. A. & Radley, J. J. Adrenocortical status predicts the degree of age-related deficits in prefrontal structural plasticity and working memory. J. Neurosci. 34, 8387-8397 (2014). 55. Allard, S., Scardochio, T., Cuello, A. C. & Ribeiro-da-Silva, A. Correlation of cognitive performance and morphological changes in neocortical pyramidal neurons in aging. Neurobiol. Aging. 33, 1466-1480 (2012). 56. Kougias, D. G. et al. Beta-hydroxy-beta-methylbutyrate ameliorates aging effects in the dendritic tree of pyramidal neurons in the medial prefrontal cortex of both male and female rats. Neurobiol. Aging. 40, 78-85 (2016). 57. Grill, J. D. & Riddle, D. R. Age-related and laminar-specific dendritic changes in the medial frontal cortex of the rat. Brain Res. 937, 8-21 (2002). 58. Nakamura, S., Akiguchi, I., Kameyama, M. & Mizuno, N. Age-related changes of pyramidal cell basal dendrites in layers III and V of human motor cortex: a quantitative Golgi study. Acta Neuropathol. 65, 281-284 (1985). 59. Pyapali, G. K. & Turner, D. A. Increased dendritic extent in hippocampal CA1 neurons from aged F344 rats. Neurobiol. Aging. 17, 601-611 (1996). 60. Flood, D. G., Buell, S. J., Defiore, C. H., Horwitz, G. J. & Coleman, P. D. Age-related dendritic growth in dentate gyrus of human brain is followed by regression in the ‘oldest old’. Brain Res. 345, 366368 (1985). 61. Buell, S. J. & Coleman, P. D. Dendritic growth in the aged human brain and failure of growth in senile dementia. Science . 206, 854-856 (1979). 62. Molnár, K. & Kéri, S. Bigger is better and worse: on the intricate relationship between hippocampal size and memory. Neuropsychologia. 56, 73-78 (2014). 63. Gonçalves, J. T. et al. In vivo imaging of dendritic pruning in dentate granule cells. Nat. Neurosci. 19, 788-791 (2016). 64. Crino, P. B. The mTOR signalling cascade: paving new roads to cure neurological disease. Nat. Rev. Neurol. 12, 379-392 (2016). 65. Tanaka, J. et al. Protein synthesis and neurotrophin-dependent structural plasticity of single dendritic spines. Science. 319, 1683-1687 (2008). 66. Fletcher, T. L., Cameron, P., De Camilli, P. & Banker, G. The distribution of synapsin I and synaptophysin in hippocampal neurons developing in culture. J. Neurosci . 11, 1617-1626 (1991).
70 67. Marrs, G. S., Green, S. H. & Dailey, M. E. Rapid formation and remodeling of postsynaptic densities in developing dendrites. Nat. Neurosci. 4, 1006-1013 (2001). 68. Tomasoni, R. et al. SNAP-25 regulates spine formation through postsynaptic binding to p140Cap. Nat. Commun. 4, 2136 (2013). 69. Klionsky, D. J. et al. Guidelines for the use and interpretation of assays for monitoring autophagy (3rd edition). Autophagy. 12, 1-222 (2016). 70. Rubinsztein, D. C., Mariño, G. & Kroemer, G. Autophagy and aging. Cell. 146, 682-695 (2011). 71. Hands, S. L., Proud, C. G. & Wyttenbach, A. mTOR's role in ageing: protein synthesis or autophagy? Aging (Albany NY). 1, 586-597 (2009). 72. Ehninger, D. et al. Reversal of learning deficits in a Tsc2+/- mouse model of tuberous sclerosis. Nat. Med. 14, 843-848 (2008). 73. Costa-Mattioli, M. & Monteggia, L. M. mTOR complexes in neurodevelopmental and neuropsychiatric disorders. Nat. Neurosci. 16, 1537-1543 (2013). 74. Ehninger, D., de Vries, P. J. & Silva, A. J. From mTOR to cognition: molecular and cellular mechanisms of cognitive impairments in tuberous sclerosis. J. Intellect. Disabil. Res. 53, 838-851 (2009). 75. Majumder, S. et al. Lifelong rapamycin administration ameliorates age-dependent cognitive deficits by reducing IL-1β and enhancing NMDA signaling. Aging Cell. 11, 326-335 (2012). 76. Halloran, J. et al. Chronic inhibition of mammalian target of rapamycin by rapamycin modulates cognitive and non-cognitive components of behavior throughout lifespan in mice. Neuroscience. 223, 102-113 (2012). 77. Kolosova, N. G. et al. Rapamycin suppresses brain aging in senescence-accelerated OXYS rats. Aging (Albany NY). 5, 474-484 (2013).
71 Table 1ǀ Results of repeated measures, t -test and two-way ANOVA on the data obtained from younger and older animals. Repeated measures Behavioral assessment (Fig. 1d,e,f) Working Memory (number of GP, BP) df F P ηρ ² Older animals (GP n= 89, BP n=87) 1,174 236.373 <0.0005 0.576 Younger animals (GP n= 63, BP n=39) 1,100 97.730 <0.0005 0.494 Reference Memory (number of GP, BP) Older animals (GP n= 99, BP n=77) 1,174 181.672 <0.0005 0.511 Younger animals (GP n= 61, BP n=41) 1,100 79.790 <0.0005 0.444 t -test Behavioral Flexibility - New Quadrant (number of GP, BP) df t P d Older animals (GP n= 62, BP n=114) 174 -18.886 <0.0005 2.882 Younger animals (GP n= 40, BP n=61) 99 13.685 <0.0005 2.738 Behavioral Flexibility - Old Quadrant (number of GP, BP) Older animals (GP n= 62, BP n=114) 158 5.047 <0.0005 0.761 Younger animals (GP n= 40, BP n=61) 99 -5.217 <0.0005 1.063 Two-way ANOVA Behavioral assessment (Fig. 1d,e,f) Performance Age Interaction df F P ηρ ² F P ηρ ² F P ηρ ² Working Memory 1,274 273.651 <0.0005 0.500 245.073 <0.0005 0.472 10.753 0.001 0.038 Reference Memory 1,274 213.618 <0.0005 0.438 238.943 <0.0005 0.466 10.141 0.002 0.036 Behavioral Flexibility New quadrant 1,273 499.625 <0.0005 0.647 6.291 0.013 0.023 0.823 0.365 0.003 Old quadrant 1,273 42.565 <0.0005 0.135 0.738 0.391 0.003 0.153 0.696 0.001 Morphological analysis - Hippocampus (Fig. 2e,i,m) Granular Neurons 1,35 3.276 0.079 0.086 67.480 <0.0005 0.658 22.030 <0.0005 0.386 CA3 pyramidal neurons (apical tree) 1,36 2.513 0.112 0.065 4.784 0.035 0.117 0.183 0.671 0.005 CA3 pyramidal neurons (basal tree) 1,36 2.586 0.116 0.067 11.326 0.002 0.239 0.763 0.388 0.021 CA1 pyramidal neurons (apical tree) 1,38 2.242 0.143 0.056 29.918 <0.0005 0.441 9.781 0.003 0.205 CA1 pyramidal neurons (basal tree) 1,38 3.356 0.075 0.081 2.045 0.161 0.051 0.554 0.461 0.014 Sholl analysis - Hippocampus (Fig. 2f,j,n) Granular Neurons 1,35 3.482 0.070 0.090 44.316 <0.0005 0.559 13.869 0.001 0.284 CA3 pyramidal neurons (apical tree) 1,36 3.022 0.091 0.077 2.254 0.142 0.059 1.049 0.313 0.028 CA1 pyramidal neurons (apical tree) 1,38 0.059 0.809 0.002 3.546 0.067 0.085 2.456 0.125 0.061 Western Blot Data - Hippocampus (Fig.3a,b,d,e,f,g) LC3-II 1,30 9.964 0.004 0.270 1.112 0.299 0.040 0.270 0.608 0.010 P62 1,30 1.101 0.303 0.039 1.545 0.225 0.054 1.930 0.176 0.067 BDNF 1,30 0.526 0.475 0.019 0.128 0.724 0.005 2.126 0.156 0.073 PSD95 1,30 1.876 0.182 0.065 2.066 0.162 0.071 1.175 0.288 0.042 Synaptophysin 1,29 1.626 0.214 0.059 0.012 0.913 0.000 2.558 0.122 0.090 SNAP25 1,30 0.595 0.447 0.022 2.202 0.149 0.075 4.944 0.035 0.155 * p<0.05; ** p<0.01;*** p<0.001.
72 Supplementary material Figure S1| Behavioral assessment of younger and older rats. Table S2| List of correlations between the performance in working memory or behavioral flexibility tasks and the dendritic length of dorsal HPC neurons. Figure S3| Cognitive performances in the WM, RM and BF tasks of the animals used in the morphological and molecular analysis. Figure S4| Morphological analysis of neurons in the mPFC. Table S5| Correlations between the performance in the behavioral flexibility task and the dendritic length of mPFC neurons. Figure S6| Morphological alterations in the HPC and mPFC of individual young and old animals. Table S7| List of correlations between dorsal HPC western blot data and the performance in the reference memory, working memory or behavioral flexibility tasks. Figure S8| Dysregulation in autophagy signaling and dendritic pruning in the mPFC of older animals. Table S9| Correlations between mPFC western blot data and the performance in the reference memory, working memory or behavioral flexibility tasks. Figure S10| Full-length western blots for Figure 3c and supplementary Figure S8c.
73 Figure S1| Behavioral assessment of younger and older rats. When similar age animals are clustered (see methods for details) in Good and Bad performers: A, B) Learning curves in the working (A) and reference memory task (B) of GPs and BPs for both younger and older rats. C) Results from the behavioral flexibility task. Average time spent on the four trials in each imaginary quadrant is given as a percentage of the total escape latency. Number of animals: working memory - older: GPs n=89, BPs n=87; younger: GPs n=63, BPs n=39; reference memory - older: GPs n=99, BPs n=77; younger: GPs n=61 BPs n=41; behavioral flexibility task - older: GPs n=62 BPs n=114; younger: GPs n=40 BPs n=61. Error bars represent SEM; * p<0.05; *** p<0.001.
74 Table S2| List of correlations between the performance in working memory or behavioral flexibility tasks and the dendritic length of dorsal HPC neurons. Old animals Young animals WM BF WM BF Granular neurons Dendritic length Pearson Correlation -0.458* 0.029 0.159 0.533 Sig. (2-tailed) 0.024 0.894 0.572 0.050 N 24 24 15 14 CA3 pyramidal neurons Apical dendritic length Pearson Correlation -0.431* 0.101 -0.230 -0.649* Sig. (2-tailed) 0.036 0.645 0.410 0.012 N 24 23 15 14 Basal dendritic length Pearson Correlation -0.123 0.060 -0.173 -0.585* Sig. (2-tailed) 0.567 0.785 0.538 0.028 N 24 23 15 14 CA1 pyramidal neurons Apical dendritic length Pearson Correlation -0.418* 0.095 0.246 0.351 Sig. (2-tailed) 0.030 0.644 0.377 0.219 N 27 26 15 14 Basal dendritic length Pearson Correlation -0.436* 0.006 0.111 -0.095 Sig. (2-tailed) 0.023 0.976 0.694 0.748 N 27 26 15 14 *p<0.05
75 Figure S3| Cognitive performances in the WM, RM and BF tasks of the animals used in the morphological and molecular analysis. A, B, C, E and F) Represent the cognitive cluster of each animal (both young and aged) used for the analysis of the Dg, CA3, CA1, Cg/PL and IL, respectively. D and G) Represent the cognitive clusters of each animal (both young and aged) used for HPC (A) and mPFC (B) western blot analysis of LC3, p62, BDNF, PSD95, SNAP25 and Synaptophysin levels. Arrows indicate missing proteins in the analysis for each animal.
82 Table S9| Correlations between mPFC western blot data and the performance in the reference memory, working memory or behavioral flexibility tasks. Old animals Young animals RM WM BF RM WM BF mPFC Autophagy LC3-II Pearson Correlation 0.387 0.445 0.003 -0.462 -0.093 0.016 Sig. (2-tailed) 0.112 0.064 0.989 0.179 0.798 0.968 N 18 18 18 10 10 10 p62 Pearson Correlation -0.504* -0.646** -0.003 -0.573 0.039 -0.444 Sig. (2-tailed) 0.028 0.003 0.991 0.083 0.914 0.231 N 19 19 19 10 10 10 Dendritic growth BDNF Pearson Correlation -0.150 -0.271 -0.203 -0.218 -0.425 -0.099 Sig. (2-tailed) 0.527 0.248 0.390 0.520 0.193 0.786 N 20 20 20 11 11 11 Synaptic markers PSD95 Pearson Correlation -0.113 -0.217 0.010 0.158 -0.020 0.702* Sig. (2-tailed) 0.654 0.386 0.968 0.643 0.954 0.024 N 18 18 18 11 11 10 Synaptophysin Pearson Correlation -0.199 -0.304 -0.330 -0.262 0.297 0.055 Sig. (2-tailed) 0.401 0.192 0.155 0.436 0.376 0.879 N 20 20 20 11 11 11 SNAP25 Pearson Correlation -0.191 -0.131 0.096 0.755** 0.449 0.142 Sig. (2-tailed) 0.434 0.593 0.697 0.007 0.166 0.696 N 19 19 19 11 11 11 *p<0.05; **p<0.01
83 Figure S10| Full length western blots. Full-length western blots for figure 3c and supplementary figure S8c.
84 Chapter III Bigger is worse: volumetric correlates of age-related changes in hippocampus and medial prefrontal cortex in the rat Mota C, Pereira das Neves S, Sousa N, Sousa JC & Cerqueira JJ Manuscript in preparation
85 Bigger is worse: volumetric correlates of age-related changes in hippocampus and medial prefrontal cortex in the rat Cristina Mota1,2, Sofia Pereira das Neves1,2, Nuno Sousa1,2, João Carlos Sousa1,2 and João José Cerqueira1,2* 1Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Braga, Portugal 2ICVS/3B’s - PT Government Associate Laboratory, Braga/Guimarães, Portugal *Corresponding author: João José Cerqueira Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal Tel +351 253604928, Fax +351 253604809 Email: [email protected]minho.pt
86 1. Abstract The world population is increasingly old, making aging research a top priority. Although human aging is generally associated with decreased cognition, several individuals retain their abilities until late in life. Understanding why this happens can help the promotion of “healthy aging”. We have previously shown that older rats possess poorer spatial learning and behavioral flexibility than younger subjects, but the degree of cognitive decline was highly variable, with some old subjects performing as well as, or even better than their younger counterparts. Given this variability, we clustered younger and older animals according to individual cognitive performance as good and bad performers. Using a battery of water maze tasks, and a detailed stereological analysis, these behavioral differences were now correlated with volumetric alterations in the hippocampus (HPC) and medial prefrontal cortex (mPFC). While in younger animals, better cognitive performance was associated with increased HPC and mPFC volumes, in the older group, better performance was associated with lower volume on both brain structures. Interestingly, while working memory performance was only related with alterations in the mPFC, the reference memory performance correlated with both hippocampal and mPFC volumes. These results, together with our previous findings relating longer dendritic trees and autophagy deficits to cognitive deficits in the older population, suggest that volumetric alterations are associated with agerelated HPC and mPFC-dependent behavior deficits. Therefore, the current work once more supports the notion that while for the younger bigger is better, for the older, smaller is definitively the best. 2. Introduction Normal aging is a process which inevitably triggers plastic and adaptive changes in the brain, which generally translate as a decline in cognitive abilities (Erickson & Barnes, 2003). However, one of the most striking characteristics of human aging is its heterogeneity (Ardila, 2007; Santos et al. , 2013). Understanding why some individuals’ cognition seems to be unaffected by age could promote the design of strategies to prevent this age-related deleterious declining. Using rats as models of cognitive aging, we recently showed that, as in humans, the profile of cognitive deficits in rodents is broad and the severity of the cognitive decline associated with aging is highly variable (Mota et al. , 2018). Furthermore, we previously showed that alterations in the dendritic length of neurons from hippocampus (HPC) and medial prefrontal cortex (mPFC) underpin the heterogeneity observed in the performance of younger and older animals, with a twist. Indeed, it seems that while for younger animals “bigger is better” for older animals “smaller is definitely better”. Moreover, we provided evidence
87 that, in older animals, dendritic length differences, and concomitant behavioral heterogeneity, can be ascribed to variations in neurotrophin levels and, more importantly, autophagic activity, leading to dendritic pruning deficits (Mota et al. , 2018). Several studies have also shown that volume alterations in the HPC and mPFC areas could be an important determinant for cognitive impairment. As it was previously reported that neuronal dendritic remodelling is directly related with volumetric alterations (Cerqueira et al. , 2005, 2007), in the present work, we hypothesized that our previous findings (Mota et al. , 2018) should translate in equivalent volumetric changes in HPC and mPFC areas. Historically, it has been presumed that brain volume loss in HPC and mPFC areas underpin age-related cognitive decline, in both humans (Jernigan et al. , 1991; Golomb et al. , 1993 Driscoll et al. , 2003, 2009; Freeman et al. , 2008; Raz et al. , 2010) and rats (Rapp et al. , 1999; Driscoll et al. , 2006; Yates et al. , 2008). However, this view has been recently brought into question. For example, in humans, some studies reported no volumetric changes in the HPC during normal aging (Sullivan et al. , 1995; Raz, 1996) and another study suggested that larger HPC are associated with less effective memory performance in healthy young adults (Molnár & Kéri, 2014). Important to highlight is that most of these studies failed to accommodate the evidence of individual heterogeneity in aging. Despite reports of age-related structural variations, little is known regarding the relationship between these variations and age-related decline in HPC and mPFC-dependent learning and memory. Therefore, in an extension of our previous work, the present study intends to fill this gap in the structure-function relationship by the parallel evaluation of cognitive performance and volume alterations in HPC and mPFC. Briefly, after cognitive characterization of both younger and older animals as good and bad performers (GPs, BPs), a detailed stereological analysis was applied to estimate the volumes of the main divisions of the HPC formation and mPFC. This data showed, once again, that bigger HPC and mPFC volumes correlate to worse cognitive performance in the older rodent population. 3. Methods 3.1. Animals All procedures were carried out in accordance with local regulations (Decreto-Lei n.º 113/2013) and European Union Directive 2010/63/EU on animal care and experimentation. Animal facilities and the people directly involved in animal experiments were certified by the Portuguese regulatory entity – DGAV (Direção-Geral de Alimentação e Veterinária). All protocols were approved by the Ethics Committee of the
88 Life and Health Sciences Research Institute (ICVS). All the male Wistar Han rats (Charles River Laboratories, Barcelona, Spain) used in the study were housed in groups of 2 and maintained under standard laboratory conditions: artificial 12h light/dark cycle (lights on from 08:00 a.m. to 08:00 p.m.); room temperature 22°C; ad libitum access to food and water. From a total of 176 old (22-24-month-old) and 102 younger (4-6-month-old) male rats, previously behaviorally characterized by Mota et al. , 2018 in a battery of water maze-based tests, a subset of 13 younger and 32 older animals were used in this study. Animals were clustered, within the original animal group, into GPs and BPs, according to their performance in each water maze-based test (Mota et al. , 2018), and subjected to stereological analysis. The remaining animals were sacrificed at different time points for several other analyses not included in the present study. All behavioral testing was conducted during the daytime. 3.2. Behavioral assessment The cognitive status of all the animals was assessed based on their performance in a series of tasks using the water maze (working and reference memory, behavioral flexibility). For more details see Mota et al. , 2018. 3.3. Histological procedures Two months after the behavioral evaluation, 32 older rats and 13 younger rats were randomly selected, deeply anesthetized with sodium pentobarbital and perfused transcardially with 4% paraformaldehyde solution for glycolmethacrylate inclusion. Whenever the histological conditions affected the final quality of the ample sections, these were excluded from de analysis therefore reducing the number of available animals for the study of some brain regions. Glycolmethacrylate inclusion Brains were removed and placed in fixative. After 4 weeks, brains were split into two hemispheres by a midsagittal section and processed for stereology, according to the procedure described previously by Keuker et al. , 2001. Briefly, they were included in glycolmethacrylate (Tecnovit 7100; Heraeus Kulzer, Werheim, Germany) and every other microtome-cut section (30µm) was then collected on a gelatinized slide, stained with Giemsa, and mounted with Entellan New (Merck, Darmstadt, Germany).
89 Structural analysis In order to ensure an unbiased analysis, slides were re-coded by the lab technician (not otherwise involved in the research) as soon as they were prepared and all stereological analyses were done blind to animal age or performance group. To minimize bias, codes were only broken after all data was collected and entered into the database. Region and layer boundaries We analyzed three areas of the mPFC: cingulate (Cg), prelimbic (PL) and infralimbic (IL) cortices. Each mPFC subregion was further divided parallel to the surface in three easily distinguishable layers (layer I, layer II and layers III-VI), based on cell packing. The third level was considered as a whole because a clear boundary between its layers could not be found in the mPFC (for more details see Cerqueira et al. , 2005). The HPC was divided in dorsal (DHPC) and ventral counterpart (VHPC) according to Pinto et al. , 2015, and was analyzed according to its main anatomical divisions: dentate gyrus (DG) (including hilus, granule cell layer, and molecular layer), cornu ammonis regions 3 and 1 (CA3 and CA1) (including strata oriens, pyramidale and radiatum). The above-mentioned regions were outlined according to the atlas of Paxinos & Watson, 1998, based on noticeable cytoarchitectural differences (Palomero-Gallagher & Zilles, 2004; Vogt et al. , 2004). Stereological procedures Volume estimations were performed using StereoInvestigator software (MicroBrightField, Williston, VT) and a camera (DXC390; Sony, Tokyo, Japan) attached to a motorized microscope (Axioplan 2; Zeiss, Oberkochen, Germany). Cavalieri’s principle (Gundersen et al. , 1988) was used to assess the volume of each region. Briefly, after starting at a random position, every 8th (for IL, PL and anterior Cg cortex), 20th (for DHPC) and 10th (for VHPC) section was used and its cross-sectional area was estimated by point counting (final magnification x112). For this, we randomly superimposed onto each area a test-point grid in which the interpoint distance, at tissue level, was 75 μm for IL layers I and II; 100 μm for IL layer IIIVI and PL level I and II;150 μm for PL layer III-VI, Cg layer I and II and the three layers of the DG; 250 μm for Cg layer III-VI and the three layers of CA3 and CA1. The volume of the region of interest was determined from the number of points that fell within its boundaries and the distance between the systematically sampled sections. The estimation of volumes of the different regions of the HPC and mPFC formation were undertaken in the right hemisphere.
90 3.4. Statistical analysis All statistical analysis was conducted in the SPSS software package version 24 (IBM corporation, Armonk, New York). After confirmation of normality and homogeneity, appropriate statistical tests were applied to the data. To facilitate direct comparisons between different tests, results of all behavioral tests were converted to a 0-100% scale, where 0% indicates worst possible performance (120s to reach the platform for the working and reference memory tests or no time in target quadrant for the behavioral flexibility task) and 100% indicates best possible performance (0s to reach the platform for the working and reference memory tests or 120s in target quadrant for the behavioral flexibility task). Also, in order to allow individual correlations between the performance of the animals in the working and reference memory test, which consist of several testing days, and structural parameters, we calculated a performance index for each test by employing the following formula: Performance index=(P1+(P3+P4)/2)/2 where Pn represents the average performance of each animal on trial n (for working memory) or day n (for reference memory); importantly, the index value can be directly read as the average performance of each animal per trial/day. Regarding the behavioral flexibility test, the percentage of time spent in the target quadrant (performance index) was used to assess the individual correlations. Clustering of animals in GPs and BPs was done using the k-means cluster analysis (for more details see Mota et al. , 2018). Comparisons between groups (younger versus older animals and good versus bad performers) were done using the two-tailed t -test (for values of a single test) or the repeatedmeasures ANOVA (for repeated testing). Two-way ANOVA was used to evaluate the impact of age and the effect of group performance in further structural data. Pearson product-moment correlations were computed between continuous variables. Measures of effect size (Cohen’s d , partial Eta-squared - or Pearson correlations - r) are presented whenever appropriate. Differences were considered to be significant if *p<0.5; **p<0.01; ***p<0.001. 4. Results 4.1. Behavioral data A k-means clustering was performed to classify younger and older animals according to their performance in working memory, reference memory or behavioral flexibility tasks. This resulted, for each test and age, in two groups of subjects (GPs and BPs) (for more details see Mota et al. , 2018). In figure 1a,b, and c the behavioral performance and group membership of the subset of the animals used for the stereological analysis is presented. A two-way ANOVA analyses of the learning curves for the
91 working and reference memory tasks (Fig. 1a,b) revealed a significant effect of age and performance in both tasks, but not an interaction between these two factors (Table 1). Regarding the behavioral flexibility test, only the time spent in the new quadrant presented a significant effect of performance, but not of age, with a significant interaction between them (Table 1). Within group comparisons showed, that for older and younger individuals, significantly different performances were observed between GPs and BPs for all the three cognitive tasks evaluated (Table 1 and Fig. 1a,b,c). Overall, in memory tests, both for older and younger adult rats, the GPs performed better than BPs (Fig. 1a,b,c). Figure 1| Performance clustering of younger and older rats according to working memory, reference memory and behavioral flexibility water maze tests. A) Learning curve in the working memory (older: GPs n=15, BPs n=17; younger: GPs n=10, BPs n=4) and B) reference memory tasks (older: GPs n=18, BPs n=14; younger: GPs n=9, BPs n=5). C) Results from the behavioral flexibility task (older: GPs n=8, BPs n=24; younger: GPs n=7, BPs n=5). Error bars represent SEM; * p<0.05; ** p<0.01; *** p<0.001. 4.2. Enlargement of the hippocampal and mPFC volume with age Older animals displayed an enlargement of the volume of the HPC and mPFC when compared to their youngster counterparts (Table 2 and Fig. 2a and 3b, respectively). As there is strong evidence supporting a functional dissociation between the DHPC and the VHPC: the DHPC being predominantly dedicated to cognitive processing whereas its ventral counterpart more implicated in emotional processing, our analysis also focused on the dorsal and ventral subdivisions of the HPC. In agreement with the data of the whole HPC volume, the volume of the DHPC was significantly increased with age (Table 2 and Fig. 2a). Interestingly, no differences were found in the volume of the VHPC between older and younger animals (Table 2 and Fig. 2a). Therefore, we next analyzed in more detail the various subregions of the
98 imaging studies have identified HPC and mPFC subfields that are selectively affected by aging, whereas volumetric alterations in these structures have been coupled to the status of spatial learning (Rasmussen et al. , 1996; Rapp et al. , 1996 and 1999; Van Petten, 2004). The implicit link for aging is, of course, an existence of volumetric atrophy which impairs function. However, this link is not as trivial as it appears. These contradictory findings could be justified by the fact that most of the studies performed in humans relay on humans with pathology and that the heterogeneity observed in the aging brain has not been a focus of attention in previous volumetric studies. Also, in contrast to the common and intuitive belief that larger HPC are better, some studies surprisingly reported a negative correlation between cognitive abilities and HPC volume (Chantôme et al. , 1999; Van Petten, 2004; Molnár & Kéri, 2014). Molnár & Kéri (2014) demonstrated that in individuals with Fragile X Syndrome with known abnormal pruning, there is an inverse correlation between HPC volume and cognitive performance: larger HPC was associated with worse general memory, which is not present in age-matched individuals without pruning deficits. In addition, a meta-analysis from Van Petten (2004) demonstrated also a negative correlation between HPC size and memory in young adults, whereas the correlation was positive in older participants. According to some authors, this paradoxical negative correlation might be related to incomplete synaptic pruning during childhood and adolescence, which refers to the elimination of unnecessary neurons and synapses to achieve more economic information processing (Foster et al. , 1999; Pohlack et al. , 2014); while other authors, in a more rudimentary way, explained this by the amount of effort provided by the subject: those who had difficulties performing the task tended to make greater effort and thus activated the associated structure(s) more, while those for whom the task was less difficult used more efficient strategies and needed less effort (Parks et al. , 1988). Hence, larger hippocampus might be less optimal for learning and memory. Our own previous findings have shown that larger dendritic trees were associated with best cognitive performances in young subjects, whereas the opposite was true for older individuals, where decreased neuronal complexity, driven by a more efficient autophagy-dependent dendritic pruning, was associated with a healthier cognitive aging process (Mota et al. , 2018). First of all, it is important to mention that the present results document an enlargement of the HPC and mPFC volumes in the older rats when compared to the youngster counterparts. These data are against the majority of studies reported so far. For instance, Hamezah and collaborators in 2017, using magnetic resonance imaging, showed that the mPFC and HPC volumes were smaller in 27-month-old rats than in 14-month-old rats.
99 Nevertheless, in accordance with our results, Ojo B. et al. , (2013) have shown that the volume of DHPC CA3 increases with age. We cannot exclude, however, that alterations in astrocytic number, axonal myelination, and extracellular volume, could influence the observed differences between young and aged individuals. Besides, our data was not normalized to the whole-brain volume as widely described in human studies (reviewed by van Petten C, 2004). Nevertheless, since our main goal was to understand the interindividual differences between younger and older animals, and between GPs and BPs within each age category, age-related volume alterations will probably not interfere with our interpretations. Thus, bearing in mind the categorization of the animals as GPs and BPs, the performance of both older and younger animals in the reference memory task (HPC-dependent task) correlates with volumetric alterations in HPC and mPFC area, while the performance in the working memory tasks (mPFC-dependent task), correlates solely with mPFC volumetric alterations. Moreover, our work revealed that young animals classified as GPs, when compared to the bad ones, presented, in general, an increased HPC and mPFC volume. Surprisingly, old animals classified as GPs showed the opposite, as they presented a decreased volume on both brain structures, whereas the bad ones presented an increased volume. These results fit with the morphological data previously reported by Mota et al. , 2018 for young and old animals and translate in equivalent volumetric changes in HPC and mPFC areas. While in young animals, better cognitive performance was associated with longer dendritic trees and higher levels of synaptic markers, this association was exactly the opposite in the older group, in which better performance was associated with shorter dendritic branches and lower levels of synaptic markers (Mota et al. , 2018). Moreover, our data further strengths the evidence that, in old animals with deficits in HPC and mPFCdependent tasks, the larger dendritic trees observed result from impaired autophagy, leading to dendritic pruning deficits and consequently impaired memory (Mota et al. , 2018). In line with this and taking into account the study of Molnár & Kéri (2014), it is plausible to suggest that the inverse correlation, between HPC volume and cognitive performance in the elderly, might be attributed to age-associated dendritic pruning deficits leading to larger, less efficient, dendritic trees and consequently increased volume. Of note, in our work, the inverse correlations between volume/dendritic tree length and cognitive performance are not present in the group of younger adults (in which an opposite trend is observed), pointing to an age-related phenomenon. Furthermore, volumetric alterations in the HPC only occurred in the dorsal pole of the HPC axis. These results are in line with the different functions of the two hippocampi: the DHPC is involved in memory and cognitive processing, whereas the VHPC processes information related to the emotional and homeostatic states of the animal (Fanselow & Dong, 2010; McHugh et al. , 2011). Given that the integrity
100 of the HPC circuitries is crucial for the numerous functions ascribed to this region of the brain, it is conceivable that structural alterations in the HPC might lead to circuit disruption and to the compromise of the successful performance in the spatial memory tasks (Morris, 1984). While the DG is thought to contribute to the formation of new memories, the CA3 plays an important role in the encoding of new spatial information in short-term memory whereas the CA1 is important for representing and remembering spatial information (Amaral et al. , 2007; Kesner et al. , 2007; Ji & Maren, 2008). Accordingly, our data showed that the volumetric alterations in young animals were observed specifically in the DG granular layer and in the stratum radiatum of CA1 both in the dorsal pole of the HPC axis. A similar result was observed for the aged animals, with alterations in the volume of the dorsal DG (molecular, granular and hilus) and dorsal CA1 (strata oriens, pyramidale and radiatum). Interestingly, no alterations in the CA3 area were observed between GPs and BPs, either in young or aged animals. Thus, it can be inferred that both the DG and CA1 were preferentially affected by age and could, therefore, be linked to alterations in spatial learning, while the CA3 region seems to be fairly resistant to the aging process. These particular data are, in one hand, in line with the research of Yang et al. , (2013) where they revealed that, unlike in CA1 synapses, the high frequency stimulation of the associative/commissural pathway leading to CA3 long-term potentiation is minimally affected by age; on the other hand, its against the majority of studies that documented a role for CA3 in the acquisition of spatial memory in the Morris water maze task (Steffenach et al. , 2002; Jo et al. , 2007). These data then suggest that age-related changes may not be uniform across the hippocampus, and given the complexity of this structure, whether changes are seen depends on the specific region under investigation. The mPFC plays a key role in decision making, planning, processing of emotional stimuli, behavioral flexibility, in social interactions, as well as in working memory (Damasio, 2000; Manes et al. , 2002; Muller et al. , 2002). In the rat, the mPFC consists of three main subdivisions which are the: Cg, PL and IL cortices (Van Eden, 1985). These different subdivisions also appear to serve separate and distinct functions. For example, the Cg has been implicated in the control of actions, decision making and plays a key role in the expression of remote memory (Teixeira et al. , 2006), while the PL and IL area have been associated with autonomic/emotional control and have remarkably dense reciprocal connections with the HPC formation, thus being specifically implicated in working memory. (Ongur & Prince, 2000; Vertes, 2004; Gisquet-Verrrier & Delatour, 2006; Cerqueira et al. , 2008; Euston et al. , 2012). Our results showed that the behavioral correlates of mPFC volume were different for younger and older animals. Specifically, in younger animals, in general, the regional volume of mPFC positively correlates with the working memory performance, while in older rats volumetric alterations negatively correlates with the reference
101 memory performance. As expected, and similar to what we previously described (Mota et al. , 2018), the direction of changes was similarly reversed between young and aged individuals, with young GPs presenting larger and aged GPs smaller mPFC volumes, when compared with BPs. The areas responsible for the differences observed between the younger GPs and BPs animals were the PL an IL areas, which is in accordance with the role of these areas regarding the working memory performance (Gisquet-Verrrier & Delatour, 2006). By contrast, in older animals volumetric alterations was essentially described in the Cg area. This area has been associated with remote memory and in conformity with this, volumetric changes of Cg area correlates with reference memory performance (long-term memory task) (Teixeira et al. , 2006). Given this, and similarly to what happens in the hippocampus, the age-related volumetric changes may be different across the mPFC. In summary, our findings show some important changes that occur in the rat HPC and mPFC with normal aging, and these changes are related selectively to different cognitive performances, regarding HPC or mPFC-dependent memory tasks. Taken together, these results strongly support that volumetric alterations in HPC and mPFC are pivotal in mediating age effects in the brain and fits with our previous data regarding neuron morphology (Mota et al. , 2018). A summary of the volumetric data is presented in figure 4. Both for volumetric and morphological data, the theory that “bigger is always better” for the performance of young animals it’s true, however, and apparently counterintuitive, for aged animals, “smaller is better”. Hopefully, our work will provide a novel hypothesis to understand the individual differences observed with aging not only in rodents but also in humans. Understanding the mechanisms behind volumetric and dendritic changes and how these changes contribute to the brain’s capacity for memory, will aid in the development of new therapeutic avenues and to prevent or restore cognitive impairments towards a successful brain aging.
102 Figure 4| Schematic representation of the volumetric alterations in the HPC and mPFC in cognitive aging. In younger animals “bigger is better”; better cognitive performance was associated with increased HPC and mPFC volumes. In older animals, it seems that “smaller is better”. BPs have increased HPC and mPFC volumes. 6. References • Amaral, D.G., Scharfman, H.E. & Lavenex, P. The dentate gyrus: fundamental neuroanatomical organization (dentate gyrus for dummies). Prog. Brain Res. 163, 3-22 (2007). • Ardila, A. Normal aging increases cognitive heterogeneity: analysis of dispersion in WAIS-III scores across age. Arch. Clin. Neuropsychol. 22, 1003-1011 (2007). • Burke, S.N. & Barnes, C.A. Neural plasticity in the ageing brain. Nat. Rev. Neurosci. 7, 30-40 (2006). • Cerqueira, J.J, Almeida, O.F. & Sousa, N. The stressed prefrontal cortex. Left? Right! Brain Behav. Immun. 22, 630-638 (2008).
103 • Cerqueira, J.J., Pêgo, J.M., Taipa, R., Bessa, J.M., Almeida, O.F.X. & Sousa, N. Morphological correlates of corticosteroid-induced changes in prefrontal cortex-dependent behaviors. J. Neurosci. 25, 7792-7800 (2005). • Cerqueira, J.J., Taipa, R., Uylings, H.B.M., Almeida, O.F.X. & Sousa, N. Specfic configuration of dendritic degeneration in pyramidal neurons of the medial prefrontal cortex induced by differing corticosteroid regimens. Cereb. Cortex. 17, 1998-2006 (2007). • Chantôme, M., Perruchet, P., Hasboun, D., Dormont, D., Sahel, M., Sourour, N., Zouaoui, A., Marsault, C. & Duyme, M. Is there a negative correlation between explicit memory and hippocampal volume? Neuroimage. 10, 589-595 (1999). • Damasio, A.R. Eighth C.U. Ariens Kappers Lecture. The fabric of the mind: a neurobiological perspective. Prog. Brain Res. 126, 457-467 (2000). • de Brabander, J.M., Kramers, R. J. & Uylings, H.B. Layer-specific dendritic regression of pyramidal cells with ageing in the human prefrontal cortex. Eur. J. Neurosci. 10, 1261-1269 (1998). • Dickstein, D.L., Kabaso, D., Rocher, A.B., Luebke, J.I., Wearne, S.L. & Hof, P.R. Changes in the structural complexity of the aged brain. Ageing Cell. 6, 275-284 (2007). • Dickstein, D.L, Weaver, C.M., Luebke, J.I. & Hof, P.R. Dendritic spine changes associated with normal aging. Neuroscience. 251, 21-32 (2013). • Driscoll, I., Davatzikos, C., An, Y., Shen, D., Kraut, M. & Resnick, S.M. Longitudinal pattern of regional brain volume change differentiates normal aging from MCI. Neurology. 72, 1906-1913 (2009). • Driscoll, I., Hamilton, D.A., Petropoulos, H., Yeo, R.A., Brooks, W.M., Baumgartner, R.N. & Sutherland, R.J. The aging hippocampus: cognitive, biochemical and structural findings. Cereb. Cortex. 13, 1344-1351 (2003). • Driscoll, I., Howard, S.R., Stone, J.C., Monfils, M.H., Tomanek, B., Brooks, W.M. & Sutherland, R.J. The aging hippocampus: a multi-level analysis in the rat. Neuroscience . 139, 1173-1185 (2006). • Erickson, C.A. & Barnes, C.A. The neurobiology of memory changes in normal aging. Exp Gerontol, 38, 61-69 (2003). • Euston, D.R., Gruber, A.J. & McNaughton, B.L. The role of medial prefrontal cortex in memory and decision making. Neuron. 76, 1057-1070 (2012). • Fanselow, M.S. & Dong, H.W. Are the dorsal and ventral hippocampus functionally distinct structures?. Neuron. 65, 7-19 (2010).
104 • Foster, J.K., Meikle, A., Goodson, G., Mayes, A.R., Howard, M., Sunram, S.I., Cezayirilli, E. & Roberts, N. The hippocampus and delayed recall: Bigger is not necessarily better? Memory. 7, 715-732 (1999). • Freeman, S.H., Kandel, R., Cruz, L., Rozkalne, A., Newell, K., Frosch, M.P., Hedley-Whyte, E.T., Locascio, J.J., Lipsitz, L.A. & Hyman, B.T. Preservation of neuronal number despite age-related cortical brain atrophy in elderly subjects without Alzheimer disease. J. Neuropathol. Exp. Neurol. 67, 1205-1212 (2008). • Gisquet-Verrier, P. & Delatour, B. The role of the rat prelimbic/infralimbic cortex in working memory: not involved in the short-term maintenance but in monitoring and processing functions. Neuroscience. 141, 585-596 (2006). • Golomb, J., Deleon, M.J., Kluger, A., George, A.E., Tarshish, C. & Ferris, S.H. Hippocampal atrophy in normal aging: an association with recent memory impairment. Arch. Neurol. 50, 967-973 (1993). • Gundersen, H.J., Bendtsen, T.F., Korbo, L., Marcussen, N., Moller, A., Nielsen, K., Nyengaard, J.R., Pakkenberg, B., Sorensen, F.B., Vesterby, A. & West, M.J. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS. 96, 379-394 (1988). • Hamezah, H.S., Durani, L.W., Ibrahim, N.F., Yanagisawa, D., Kato, T., Shiino, A., Tanaka, S., Damanhuri, H.A., Ngahb, W.Z.W. & Tooyama, I. Volumetric changes in the aging rat brain and its impact on cognitive and locomotor functions. Exp. Gerontol. 99, 69-79 (2017). • Jernigan, T.L., Archibald, S.L., Berhow, M.T., Sowell, E.R., Foster, D.S., Hesselink, J.R. Cerebral structure on MRI, part I. Localization of age-related changes. Biol. Psychiatry. 29, 55-67 (1991). • Ji, J. & Maren, S. Differential roles for hippocampal areas CA1 and CA3 in the contextual encoding and retrieval of extinguished fear. Learn. Mem. 15, 244-251 (2008). • Jo, Y.S., Park, E.H., Kim, I.H., Park, S.K., Kim, H., Kim, H.T. & Choi, J.S. The medial prefrontal cortex is involved in spatial memory retrieval under partial-cue conditions. J. Neurosci. 27, 13567-13578 (2007). • Kesner, R.P. Behavioral functions of the CA3 subregion of the hippocampus. Learn. Mem. 14, 771781 (2007). • Keuker, J.I., Vollmann-Honsdorf, G.K. & Fuchs, E. How to use the optical fractionator: an example based on the estimation of neurons in the hippocampal CA1 and CA3 regions of tree shrews. Brain Res. Brain Res. Protoc. 7, 211-221 (2001).
105 • Manes, F., Sahakian, B., Clark, L., Rogers, R., Antoun, N., Aitken, M. & Robbins, T. Decision-making processes following damage to the prefrontal cortex. Brain. 125, 624-639 (2002). • Markham, J.A. & Juraska, J.M. Aging and sex influence the anatomy of the rat anterior cingulate cortex. Neurobiol. Aging. 23, 579-588 (2002). • McHugh, S.B., Fillenz, M., Lowry, J.P., Rawlins, J.N. & Bannerman, D.M. Brain tissue oxygen amperometry in behaving rats demonstrates functional dissociation of dorsal and ventral hippocampus during spatial processing and anxiety. Eur. J. Neurosci. 33, 322-337 (2011). • Merrill, D.A., Chiba, A.A. & Tuszynski, M.H. Conservation of neuronal number and size in the entorhinal cortex of behaviorally characterized aged rats." J. Comp. Neurol. 438, 445-456 (2001). • Molnár, K. & Kéri, S. Bigger is better and worse: on the intricate relationship between hippocampal size and memory. Neuropsychologia. 56, 73-78 (2014). • Morris, R. Developments of a water-maze procedure for studying spatial learning in the rat. J. Neurosci. Methods. 11, 47-60 (1984). • Mota, C., Taipa, R., Pereira das Neves, S., Monteiro-Martins, S., Monteiro, S., Palha, J.A., Sousa, N., Sousa, J.C. & Cerqueira, J.J. Structural and molecular correlates of cognitive aging in the rat. Manuscript accepted for publication in Scientific Reports, 2018. • Muller, N.G., Machado, L. & Knight, R.T. Contributions of subregions of the prefrontal cortex to working memory: evidence from brain lesions in humans. J. Cogn. Neurosci. 14, 673-686 (2002). • Ojo, B., Davies, H., Rezaie, P., Gabbott, P., Colyer, F., Kraev, I. & Stewart, M.G. (2013) “Age-induced loss of mossy fibre synapses on CA3 thorns in the CA3 stratum lucidum.” Neurosci. J. 2013, 839535; 10.1155/2013/839535. (2013). • Ongur, D. & Price, J.L. The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb. Cortex. 10, 206-219 (2000). • Palomero-Gallagher, N. & Zilles, K. Isocortex in The Rat Nervous System (ed. Paxinos, G.) 729-757 (Academic, 2004). • Parks, R.W., Loewenstein, D.A., Dodrill, K.L., Barker, W.W., Yoshii, F., Chang, J.Y., Emran, A., Apicella, A., Sheramata, W.A. & Duara, R. Cerebral metabolic effects of a verbal fluency test: A PETscan study. J. Clin. Exp. Neuropsychol. 10, 565-575 (1988). • Paxinos, G. & Watson, C. The Rat Brain In Stereotaxic Coordinates (ed. Paxinos, G.) (Academic, 1998).
106 • Pinto, V., Costa, J.C., Morgado, P., Mota, C., Miranda, A., Bravo, F.V., Oliveira, T.G., Cerqueira, J.J. & Sousa, N. Differential impact of chronic stress along the hippocampal dorsal-ventral axis. Brain Struct. Funct. 220, 1205-1212 (2015). • Pohlack, S.T., Meyer, P., Cacciaglia, R., Liebscher, C., Ridder, S. & Flor, H. Bigger is better! Hippocampal volume and declarative memory performance in healthy young men. Brain Struct. Funct. 219, 255-267 (2014). • Rapp, P. R. & Gallagher, M. Preserved neuron number in the hippocampus of aged rats with spatial learning deficits. Proc. Natl. Acad. Sci. USA. 93, 9926-9930 (1996). • Rapp, P.R., Stack, E.C. & Gallagher, M. Morphometric studies of the aged hippocampus: I. Volumetric analysis in behaviorally characterized rats. J. Comp. Neurol. 403, 459-470 (1999). • Rasmussen, T., Schliemann, T., Sørensen, J.C., Zimmer, J. & West, M.J. Memory impaired aged rats: no loss of principal hippocampal and subicular neurons. Neurobiol. Aging. 17, 143-147 (1996). • Raz, N. Neuroanatomy of aging brain: evidence from structural MRI in Neuroimaging II. Clinical Applications (ed. Bigler, E.D.) 153-182 (Academic Press,1996). • Raz, N., Ghisletta, P., Rodrigue, K. M., Kennedy, K. M. & Lindenberger, U. Trajectories of brain aging in middle-aged and older adults: regional and individual differences. Neuroimage. 51, 501-511 (2010). • Santos, N.C., Costa, P.S., Cunha, P., Cotter, J., Sampaio, A., Zihl, J., Almeida, O.F.X., Cerqueira, J.J., Palha, J.A. & Sousa, N. Mood is a key determinant of cognitive performance in communitydwelling older adults: a cross-sectional analysis. Age. 35, 1983-1993 (2013). • Steffenach, H.A., Sloviter, R.S., Moser, E.I. & Moser, M.B. Impaired retention of spatial memory after transection of longitudinally oriented axons of hippocampal CA3 pyramidal cells. Proc. Natl. Acad. Sci. USA. 99, 3194-3198 (2002). • Sullivan, E.V., Marsh, L., Mathalon, D.H., Lim, K.O. & Pfefferbaum, A. Age-related decline in MRI volumes but of temporal lobe gray matter but not hippocampus. Neurobiol. Aging. 16, 591-606 (1995). • Teixeira, C.M., Pomedli, S.R., Maei, H.R., Kee, N. & Frankland, P.W. Inolvement of the anterior cingulate cortex in the expression of remote spatial memory. J. Neurosci. 26, 7555-7564 (2006). • Van Eden, C.G. & Uylings, H.B.M. Cytoarchitectonic development of the prefrontal cortex in the rat. J. Comp. Neurol. 241, 253-67 (1985). • Van Petten, C. Relationship between hippocampal volume and memory ability in healthy individuals across the lifespan: review and meta-analysis. Neuropsychologia . 42, 1394-1413 (2004).
107 • Vertes, R.P. Differential projections of the infralimbic and prelimbic cortex in the rat. Synapse. 51, 32-58 (2004). • Vogt, B.A., Vogt, L. & Farber, N. Cingulate cortex and disease models in The Rat Nervous System (ed. Paxinos, G.) 705-727 (Academic, 2004). • Yang, S., Megill, A., Ardiles, A.O., Ransom, S., Tran, T., Koh, M.T., Lee, H.K., Gallagher, M. & Kirkwood, A. Integrity of mGluR-LTD in the associative/commissural inputs to CA3 correlates with successful aging in rats. J. Neurosci. 33, 12670-12678 (2013). • Yates, M.A., Markham, J.A., Anderson, S.E., Morris, J.R. & Juraska, J.M. Regional variability in agerelated loss of neurons from the primary visual cortex and medial prefrontal cortex of male and female rats. Brain Res. 1218, 1-12 (2008).
114 Chapter IV Adulthood cognitive training selectively prevents age-related memory impairments in the rat Mota C, Pinto V, Sousa N, Sousa JC & Cerqueira JJ Manuscript in preparation
115 Adulthood cognitive training selectively prevents aged-related memory impairments in the rat Cristina Mota1,2, Vitor Pinto1,2, Nuno Sousa1,2, João Carlos Sousa1,2 and João José Cerqueira1,2* 1Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Braga, Portugal 2ICVS/3B’s - PT Government Associate Laboratory, Braga/Guimarães, Portugal *Corresponding author: João José Cerqueira Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal Tel +351 253604928, Fax +351 253604809 Email: [email protected]minho.pt
116 1. Abstract In general, old rats present memory impairments. Since cognitive training can enhance memory function, here we tested if it is also able to prevent age-associated deficits. The Hole board (HB) test was used as a cognitive-training tool for hippocampal (HPC)-dependent spatial reference memory, and animals were trained in four sets of training paradigms, designed to assess the effects of cognitive training, but also to explore the influence of variables such as the optimal time to start the training task. Exposure to only one training session at 10 months of age, but not at 15 or 20 months, was sufficient to improve reference memory performance after one year. By contrast, this training paradigm was unable to induce any change in working memory or behavioral flexibility skills of the same animals, suggesting that its effects are task specific. When we analyzed the structural HPC alterations underpinning these behavioral improvements, differences were only detected when the animals were divided into good and bad performers (GPs, BPs). Under these analysis conditions, the group of BPs untrained old animals had the biggest HPC and the longest DG and CA1 apical dendritic trees, in line with our previous observations that poor cognitive performance in the old age was associated with pruning deficits and longer dendritic arborizations. In conclusion, these results suggest that cognitive training is long lasting in time, mainly triggering synaptic plasticity within the brain circuits involved in the specific training paradigm. Importantly, this study suggests that even a relatively short period of cognitive enrichment is sufficient to improve the ability to recover from age-induced impairments, which might prove relevant for successful aging. These promising findings underline the existence of continuous functional plasticity, which brings optimism about the possibility of improving cognitive function in old age. 2. Introduction The aging process presents a great inter-individual variability. Nevertheless, it is generally accepted that aging is mostly associated with decreasing cognition (Erickson & Barnes, 2003; Singh-Manoux et al. , 2012; Santos et al. , 2013). Since memory impairments impose a substantial burden for those affected, the development of interventions to maintain and optimize cognitive functioning has skyrocketed. Several studies have tried to tackle on strategies to prevent cognitive decline in old age, suggesting interventions based on nutritional improvements (Richards et al. , 2002), physical exercise (Tanigawa et al. , 2014) and cognitive training (Nouchi et al. , 2012; Jiang et al. , 2016). In humans, cognitive training, both early and later in life, through engagement in intellectually stimulating activities, is associated with better cognitive functioning, as it postpones or attenuates cognitive decline,
117 and ameliorates cognitive deficits (Wilson et al. , 2002; reviewed by Frick & Benoit, 2010 and Milgram et al. , 2006; Belleville & Bherer, 2012; Rebok et al. , 2014). These effects could be explained by the cognitive reserve hypothesis, which rests on the ability of the brain to compensate for pathological changes associated with aging, depending on the previous stage of intellectual capability (Whalley et al. , 2004). As such, cognitive training induces increased dendritic length of neurons (Jacobs et al. , 1993), increased gray matter volumes in the hippocampus (HPC) (Maguire et al. , 2006; Cannonieri et al. , 2007), increased cortical and subcortical volumes (Seider et al. , 2016) and resulted in important changes of brain activity (Hempel et al. , 2004; Olesen et al. , 2004). On the other hand, in rodents, the beneficial effects of cognitive training have also been associated with physical stimulation, which together is often referred to as environmental enrichment (EE) (Fischer, 2016). Environmentally enriched rodents are typically socially housed in large groups and exposed to a variety of stimulating objects that can provide both cognitive stimulation (e.g., toys, tunnels, dwellings) and physical exercise (e.g., running wheels). For example, Vicens et al. , (1999, 2002) found that a water maze training task, performed in 6-month-old mice, led to an improved performance in the water maze later at 10 and 18 months of age. More recently, Galeano et al. , 2015 showed that life-long EE was able to rescue memory deficits in control aged rats and in aged rats that were subjected to asphyxia at birth. Additionally, several research groups have systematically shown that EE induces synaptogenesis (Greenough & Volkmar, 1973; Rampon et al. , 2000), neurogenesis (Kempermann et al. , 2002), cortical thickening (Mohammed et al. , 2002) and dendritic branching (Faherty et al. , 2003; Kolb et al. , 2003; Bindu et al. , 2007). Interestingly, these anatomical changes correlate with functional alterations such as increased HPC long-term potentiation (LTP) (Artola et al. , 2006, Stein et al. , 2016). Since the HPC is particularly vulnerable to aging (Mora et al. , 2007; Mota et al. , 2018), and the Hole Board (HB) task is a highly demanding task which works as a stimulus for spatial reference memory (HPC-dependent task) (Van der Staay, 1999; Depoortère et al. , 2010), we hypothesized that the HB task would have the ability to enhance memory function in old animals. To this end, cognitive training was performed in rats of different ages that were later assessed for cognitive performance in a battery of water maze tests. In addition, we tested if there was a relationship between HPC-dependent HB training and changes in HPC morphology and neuron number.
118 3. Methods 3.1. Animals All procedures were carried out in accordance with local regulations (Decreto-Lei n.º 113/2013) and European Union Directive 2010/63/EU on animal care and experimentation. Animal facilities and the people directly involved in animal experiments were certified by the Portuguese regulatory entity – DGAV (Direção-Geral de Alimentação e Veterinária). All protocols were approved by the Ethics Committee of the Life and Health Sciences Research Institute (ICVS). All the male Wistar Han rats (Charles River Laboratories, Barcelona, Spain) used in the study were housed in groups of 2 and maintained under standard laboratory conditions: artificial 12h light/dark cycle (lights on from 8:00 a.m. to 8:00 p.m.); room temperature 22°C; ad libitum access to food and water. All behavioral testing was conducted during the daytime. A total of 82 male rats were used in this study divided in 3 sets of experiments, with the only difference between them being the age at which cognitive training was administered (10 months, 15 months, 20 months). Animals of each experiment were randomly allocated to the group that performed cognitive training (HB group: 10 months n=22, 15 months n=9, 20 months n=11) or the control group that was put in the hole board apparatus and rewarded independently of any spatial learning (pseudoHole Board group: 10 months n=19, 15 months n=12, 20 months n=9). After training, all animals waited to reach 22 months of age when they were behaviorally characterized in a battery of water-maze based tests to assess cognition. Two months later (24 months), animals were sacrificed and their brains removed for further analysis. Brains of a subset of control and cognitively trained old animals, from the 10 months experimented exclusively, were used to perform morphological analysis (3D neuron reconstruction) and stereological analysis (volumes estimations and neuronal number). As there were no differences in the behavior of animals trained at other ages, their brains were not processed for the purpose of the present study and results will not, therefore, be presented. Details on the experimental design are depicted in figure 1.
119 Figure 1| Experimental design and timeline for behavioral tests performed by Pseudo-HB and HB groups. (mo: months). 3.2. Behavioral assessment Cognitive training - HB Task The HB food-retrieval task is a highly demanding task and works as a stimulus for spatial reference memory (HPC-dependent task). This task allows the assessment of working (WM) and reference memory (RM) performance. The HB apparatus consisted of an open arena (square board), 70x70cm, with opaque walls and 16 holes of 3.5cm diameter (Fig. 1). Four of these holes were baited with rewards (Cheerio®- like cereals) with constant trial-to-trial disposition for each animal, but random distribution patterns between different animals. To prevent the animals from using scent to find rewards, a few cereals were ground and scattered over the rest of the holes. Two days before the beginning of the task and during the 5 days of the task, the animals entered a regimen of food deprivation having access to food only 1h per day. One day before the task, a reward cereal was given to the animals so that they could get used to its smell and taste. Daily sessions of 1 trial per day were given to each animal for 5 consecutive days. At the beginning of each trial, the animal was placed on the center of the platform and allowed to explore. The trial ended whenever the animal found the 4 rewards, or after a maximum of 30min. The number of right answers (nose poke on the holes that contained rewards), wrong answers (WA) (nose poke on empty holes), eaten rewards, repetitions after reward (RAR) (nose poke on correct holes that no longer had the reward), time to find the first reward and the total time of the trial were registered, for each animal and each trial. WM errors, a performance directly related to WM, corresponded to the number of RAR and RM errors, a measure related to RM status, were given by the number of WA. The pseudo-HB group was subjected to the same experimental conditions except that the pellets were exposed in the open arena.
120 Water maze tests The cognitive status of all animals was assessed based on their performance in a series of tasks using the water maze. Animals were tested during 8 days in 3 tests designed to assess different cognitive domains: WM, RM and behavioral flexibility (BF) (Cerqueira et al. , 2007). The apparatus consisted of a large circular black pool (170 cm diameter), filled to a depth of 31 cm with water (at 22°C), which was divided in 4 equal-sized quadrants by imaginary lines. During the execution of the test, a submerged cylindrical black platform (12 cm diameter, 30 cm high) was hidden 2 cm below the water surface at the center of one of the quadrants. The room was dimly lit and extrinsic visual clues were glued to the walls surrounding the tank and kept unaltered during the duration of the experiment. Data was collected using a video camera placed above the center of the pool connected to a video-tracking system (Viewpoint, Champagne au Mont d’Or, France). Working memory task: this task is a variation of the spatial RM test (Morris, 1984) and depends on medial pre-frontal cortex (mPFC) function (Kesner, 2000). Its goal is to assess the ability of rats to learn the position of a hidden platform and to keep this information online during four consecutive trials. This test consisted of 4 days of acquisition in which the position of the platform was kept constant during the four daily trials [with a maximum of 120 seconds (s) per trial] but was altered to a different quadrant every day (such that all four quadrants are used). Thus, the animal cannot know where the platform was hidden on trial 1 of each day. Test sessions begun with rats facing the wall of the maze, being placed at one of the four different starting points (north, east, south, or west) which were different for each of the four daily trials. A trial was considered complete once the escape platform had been reached by the rat. Animals were then allowed to spend 30 s in the platform, after which they were towel-dried and allowed to rest in a holding cage some seconds before being returned to the maze. When the escape platform was not reached within 120 s, the experimenter guided the animal to the platform and an escape latency of 120 s was recorded. The length of the path described (distance swam) and the time spent to reach the platform (escaped latency) were recorded in the consecutive trials. Reference memory task: this test is a HPC-dependent task (Morris, 1984) that evaluates the ability of the animal to learn the location of a hidden platform during four consecutive days – spatial reference memory. After working memory assessment (on days 5-7), the platform remained in the same quadrant as on the previous day (day 4) and animals were tested for an additional 3 days of tests. The remaining procedures were similar to the already described. Behavioral flexibility task: this is a mPFC-dependent assessment and was performed after the completion of the RM task (day 8). For this test, the escape platform was moved and located in the
121 opposite quadrant of its previous 4 days location. All the procedures were similar to those described above. For this task, the time spent swimming in each quadrant was recorded and analyzed. 3.3. Histological procedures Two months after all behavioral evaluations (Mota et al. , 2018), 28 aged rats trained at 10 months (Pseudo HB n=15, HB n=13) were randomly selected, deeply anesthetized with sodium pentobarbital and perfused with saline. The left brain hemisphere was used for Golgi-Cox staining, while the right hemisphere was fixed in 4% paraformaldehyde solution for glycolmethacrylate inclusion. The remaining animals were sacrificed for other analyses not included in the present study. In order to ensure an unbiased analysis, slides resulting from all histological procedures were re-coded by the lab technician (not otherwise involved in the research) as soon as they were prepared. To minimize bias, codes were only broken after all data was collected and entered into the database. Golgi-cox staining After perfusion, brains were removed, immersed in 25 mL Golgi-Cox solution (Gibb & Kolb, 1998) (a 1:1 solution of 5% potassium dichromate and 5% mercury chloride diluted 4:10 with 5% potassium chromate) and kept in the dark at room temperature for 14 days. Brains were then transferred to a 30% sucrose solution. At this moment, brains were stored in the dark at 4ºC from a minimum of 3 days to a maximum of 2 months, before being cut on a vibratome. Coronal sections (200 µm thick) were collected in 6% sucrose and blotted dry onto cleaned, gelatin-coated microscope slides. Subsequently, sections were alkalinized in 18.7% ammonia, developed in Dektol (Kodak), fixed in Kodak Rapid Fix, dehydrated through a graded series of ethanol of increasing concentrations and cleared in xylene before being covered in mounting media (Entellan New) and coverslipped. The slides were stored in the dark and exposed to the air, at room temperature, until being analyzed. Dendritic arborizations were analyzed in the dentate gyrus (DG), cornus ammonis 3 and 1 (CA3 and CA1) of the dorsal hippocampus (DHPC). Dorsal hippocampal formation was identified according to the division described by Pinto et al. , 2015. The granular neurons of the DG were readily identified based on their round cell bodies, which were located in the stratum granulosum of the suprapyramidal and infrapyramidal blades. Pyramidal neurons from the DHPC (CA3 and CA1) were readily identified by their characteristic triangular soma shape, apical dendritic extension toward the pial surface and numerous dendritic spines. All neurons were chosen for reconstruction based on the criteria described by Uylings et al. , (1986): (i) full impregnation of the neurons along the entire length of the dendritic tree; (ii) apical
122 dendrite without truncated branches, except on the most superficial layer; (iii) presence of at least 3 primary basal dendritic shafts, each of which branched at least once (when applicable); (iv) relative isolation from neighboring impregnated cells that could interfere with analysis (clear somatic boundaries) (v) no morphological changes attributable to incomplete dendritic impregnation of Golgi-Cox staining. To minimize selection bias, slices containing the region of interest were randomly searched and the first 510 neurons fulfilling the above criteria (maximum of 3 neurons per slice) were chosen. For each selected neuron, all branches of the dendritic tree were reconstructed at 600× magnification, using a motorized microscope (Olympus BX51 Microscope with oil-objectives), attached to a camera (QImaging® Retiga2000R digital camera, Surrey, Canada) and equipped with Neurolucida software (Microbrightfield, VT, USA). A 3-D analysis of the reconstructed neurons was performed using NeuroExplorer software (Microbrightfield, VT, USA). Dendritic morphology was examined by assessing the total dendritic length. In addition, to assess differences in the arrangement of dendritic material, a 3-D version of a Sholl analysis (Uylings & Van Pelt, 2002) was performed. For this, the number of intersections of dendrites with concentric spheres positioned at radial intervals of 20µm from the soma was recorded. Glycolmethacrylate inclusion Brains were removed and placed in fixative. After 4 weeks, brains were processed for stereology, according to the procedure described previously by Keuker et al. , (2001). Briefly, they were included in glycolmethacrylate (Tecnovit 7100; Heraeus Kulzer, Werheim, Germany) and every other microtome-cut section (30 µm) was then collected on a gelatinized slide, stained with Giemsa, and mounted with Entellan New (Merck, Darmstadt, Germany). The DHPC was analyzed according to its main anatomical divisions: DG (including hilus, granule cell layer and molecular layer), CA3 and CA1 (strata oriens, pyramidale and radiatum) (for more details see Cerqueira et al. , 2005; Pinto et al. , 2015). The above-mentioned regions were outlined according to the atlas of Paxinos & Watson (1998), based on noticeable cytoarchitectural differences (Palomero-Gallagher & Zilles, 2004; Vogt et al. , 2004). Volume estimations were performed using StereoInvestigator software (MicroBrightField, Williston, VT) and a camera (DXC390; Sony, Tokyo, Japan) attached to a motorized microscope (Axioplan 2; Zeiss, Oberkochen, Germany). Cavalieri’s principle (Gundersen et al. , 1988) was used to assess the volume of each region, using a 4x lens. Briefly, after starting at a random position, every 20th section was used and its cross-sectional area was estimated by point counting (final magnification x112). For this, we randomly superimposed onto each area a test-point grid in which the interpoint distance, at tissue level, was 150 μm for all the three layers of the DG; 250 μm for the three layers of CA3 and CA1. The volume of the
123 region of interest was determined from the number of points that fell within its boundaries and the distance between the systematically sampled sections. Average cell numbers were estimated at 600× magnification using the optical fractionator method, as described previously (West et al. , 1991). Briefly, beginning at a random starting position, a grid of virtual, equally-spaced 3D-boxes (30×30×15µm for CA3 and CA1, and 20×20×15µm for DG) (same grid spacing as for volume estimations) was superimposed on every 20th section; and neurons were counted whenever their nucleus came into focus within the counting box. Neurons were differentiated from other cells on the basis of nuclear size (larger in neurons than in glia cells), a prominent nucleolus, and the shape of their perikarya attributable to dendritic emergence (Peinado et al. , 1997). 3.4. Statistical analysis All statistical analyses were conducted in the SPSS software package version 24 (IBM corporation, Armonk, New York). After confirmation of normality and homogeneity, appropriate statistical tests were applied to the data. To facilitate direct comparisons between different water maze tests, results of all water maze tasks were converted to a 0-100 % scale, where 0 % indicates worst possible performance (120s to reach the platform for WM and RM tests, or no time in target quadrant for the behavioral flexibility task) and 100% indicates best possible performance (0 s to reach the platform or total time in the target quadrant). HB performance was quantified according to WM or RM errors. Clustering of animals in GPs and BPs was done using the k-means cluster analysis according to each animal performance in the last day of the water maze RM task. This performance is considered the best performance achieved during the completion of the task. Comparisons between Pseudo-HB and HB animals were done using two-tailed t-test. Repeated measures ANOVA was used to evaluate the impact of training in the HB task and in the water maze-based tests. In addition, to test the association of cognitive training and water-maze performance in hippocampal morphology, two-way ANOVAs were performed. Measures of effect size (Cohen’s d or Eta-squared) are presented whenever appropriate. Differences were considered significant if *p<0.5; **p<0.01; ***p<0.001. 4. Results 4.1. Cognitive training All animal groups successfully learned the HB task, as observed by the sustained decrease in the total time to perform the test (10 mo: F (2,53)=9.263, p <0.0005, ηρ²=0.306; 15 mo: F (2,16)=3.953, p =0.049,
130 behavior, the animal must learn the spatial location of the goal object relative to several distal cues (Morris, 1981). In our study, by assessing spatial learning across repeated unchanged trials, we were aiming to engage trial-independent memory performance, thus training spatial reference memory, with minor engagement of working memory skills (Van der Staay et al. , 2012). Our work revealed that exposure to cognitive training in adulthood (10 months of age) prevented reference memory impairments later on. By contrast, no changes in working memory and behavioral flexibility skills were observed. These observations strengthen the view that training on specific “cognitive tasks” may serve to reinforce or reactivate circuits and therefore to (at least partially) restore some of the age-induced deficits in cognition. In agreement with this data, Harburger et al. , 2007 reported that all enrichment treatments, including cognitive stimulation, improved spatial memory in aged females, indicating that either exercise or cognitive stimulation can improve memory in aged subjects. Interestingly, our work showed that the effects of a short period of cognitive training (1-week protocol) are long lasting in time. This points out that engagement in a highly cognitive demanding task, even for a relatively short period of time, is able to enhance cognitive function. In accordance, Arai et al. , 2009 demonstrated that when 15-day old mice were subjected to 2 weeks of EE they exhibited enhanced hippocampal LTP. Later, when the same mice were breed and LTP was analyzed in their offspring, now reared in standard cages, hippocampal LTP was enhanced when compared to offsprings from non-enriched parents. Similarly, Cheng et al. , 2012 showed that in old adults, the effects of interventions on cognition are maintained 1 year after the training has ended. Taken together, these findings could be explained by the cognitive reserve hypothesis, which rests on the ability of the brain to compensate for pathological changes associated with aging, depending on the previous stage of intellectual capability (Whalley et al. , 2004). This means that people with greater cognitive reserve can tolerate more the neurodegenerative brain changes associated with dementia or other brain pathologies, such as Parkinson's disease, multiple sclerosis, or stroke (Stern, 2012). Likewise, in our work, it seems that aged rats with higher cognitive reserve, as a consequence of the exposure to the HB test, cope better with age-related deterioration than the aged counterparts with lower cognitive reserve. Therefore, an important goal of aging research should be the promotion and sustainment of elderly people cognitive reserves. Furthermore, our findings highlight the importance of the age at which individuals are exposed to cognitive stimulation (reviewed by Frick, 2010). The influence of age to the exposure to EE seems to be controversial. Most studies indicate that EE at young ages appears to have even more beneficial effects which is most likely due to the fact that brain development is not complete and more-long lasting
131 molecular and anatomical changes are induced (Bouet et al. , 2011; Freret et al. , 2012). However, other studies found positive effects when EE training was initiated in already aged and cognitively impaired rodents (Bennett et al. , 2006). In accordance with this data, we also reported that 10 months of age seems to be the best sensitive period for the effectiveness of cognitive training, however we did not test animals younger than this time point. Our goal was also to address the underlying mechanisms of cognitive training. It is known that cognitive stimulation induces structural and functional alterations in the brain (Kolb et al. , 2003; Bindu et al. , 2007; Artola et al. , 2006, Freret et al. , 2012). Accordingly, a large body of literature has shown that exposure to an enriched environment enhances dendritic branching in rats (Greenough & Volkmar et al. , 1973; Green et al. ; 1983; Kolb et al. , 2003; Bindu et al. , 2007), as well as in humans (Jacobs et al. , 1993). However, some studies failed to see such differences (Diamond et al. , 1976), while others reported differences in dendritic branching in dentate granule cells, but only in female EE rats (Juraska et al. , 1985, 1989). Regarding volumetric changes, an increase in cortical thickness in both rats (Mohammed et al. , 2002) and healthy elderly individuals (Jiang et al. , 2016; Seider et al. , 2016) was seen after cognitive training, while another study also reported that aerobic exercise reversed age-related decreases in hippocampal volume, which correlated with an improvement in spatial skill (Erickson et al. , 2011). In the present work, structural analysis confirmed a change in neuronal length and HPC volume of worse HB performing animals that reached the level of the best performing group. Also, no significant effects of training were reported for the dendritic length or HPC volume. These findings are in contradiction with the majority of studies using EE animal models and cognitive training in humans, which report that exposure to an enriched environment enhances dendritic branching (Greenough & Volkmar et al. , 1973; Green et al. , 1983; Kolb et al. , 2003; Bindu et al. , 2007; Jacobs et al. , 1993). Of note, most of these structural studies were performed in adult rats which made it difficult to compare with our own data. Nevertheless, our results are in accordance with our previous data showing that in aged individuals “smaller is better” (Mota et al. , 2018), as the cognitively trained BP group reached the level of the best performing animals. Accordingly, aged GP untrained individuals had smaller DG and CA1 dendritic trees and smaller HPC volume when compared to BP untrained animals. Further adding to these findings, an increased HPC neuron number was observed in the cognitivelytrained group. This finding could imply that, in these animals, the aged brain is supplemented with more neurons simply as a result of new learning. Indeed, several studies reported EE induces neurogenesis (Greenough & Volkmar, 1973; Faherty et al. , 2003; Rampon et al. , 2000). Also, it´s known that HPC neurogenesis is strongly improved by physical activity, especially aerobic exercise (van Praag et al. , 1999;
132 Steiner et al. , 2008). In contrast, mental training via skill learning increases the number of surviving neurons, particularly when the training goals are challenging (Gould et al. , 1999; Shors; et al., 2012; Wurm et al ., 2007). Therefore, and bearing in mind that future work should specifically tackle this question, we hypothesize that an increase in cell survival should be the principal mechanism driving the increased neuron number. In conclusion, our work suggests that the HB test is a useful tool to enhance cognitive function, and its effects are circuity-specific and long lasting in time. The beneficial effects of cognitive training are potentially mediated by alterations in dendritic branching. These promising findings highlight the existence of continuous functional plasticity, which brings optimism about the possibility of promoting “mindspan”. 6. References • Arai, J.A., Li, S., Hartley, D.M. & Feig, L.A. Transgenerational rescue of a genetic defect 426 in longterm potentiation and memory formation by juvenile enrichment. J. Neurosci. 29, 1496-1502 (2009). • Artola, A., von Frijtag, J.C., Fermont, P.C., Gispen, W.H., Schrama, L.H., Kamal, A. & Spruijt, B.M. Long-lasting modulation of the induction of LTD and LTP in rat hippocampal CA1 by behavioural stress and environmental enrichment. Eur. J. Neurosci. 23, 261-272 (2006). • Belleville, S. & Bherer, L. Biomarkers of cognitive training effects in aging. Curr. Transl. Geriatr. Exp. Gerontol. Rep. 1,104-110 (2012). • Bennett, J.C., McRae, P.A., Levy, L.J. & Frick, K.M. Long-term continuous, but not daily, environmental enrichment reduces spatial memory decline in aged male mice. Neurobiol. Learn. Mem. 85, 139-152 (2006). • Bindu, B., Alladi, P.A., Mansooralikhan, B.M., Srikumar, B.N., Raju, T.R. & Kutty, B.M. Short-term exposure to an enriched environment enhances dendritic branching but not brain-derived neurotrophic factor expression in the hippocampus of rats with ventral subicular lesions. Neuroscience. 144, 412-423 (2007). • Bouet, V., Freret, T., Dutar, P., Billard, J. M., & Boulouard, M. Continuous enriched environment improves learning and memory in adult NMRI mice through theta burst-related-LTP independent mechanisms but is not efficient in advanced aged animals. Mech. Ageing Dev. 132, 240-248 (2011). • Cannonieri, G.C., Bonilha, L., Fernandes, P.T., Cendes, F. & Li, L.M. Practice and perfect: length of training and structural brain changes in experienced typists. Neuroreport. 18, 1063-1066 (2007).
133 • Cerqueira, J.J., Mailliet, F., Almeida, O.F., Jay, T.M. & Sousa, N. The prefrontal cortex as a key target of the maladaptive response to stress. J. Neurosci. 27, 2781-2787 (2007). • Cerqueira, J.J., Pêgo, J.M., Taipa, R., Bessa, J.M., Almeida, O.F.X. & Sousa, N. Morphological correlates of corticosteroid-induced changes in prefrontal cortex-dependent behaviors. J. Neurosci. 25, 7792-7800 (2005). • Cheng, Y., Wu, W., Feng, W., Wang, J., Chen, Y., Shen, Y., Li, Q. Zhang, X. & Li, C. The effects of multi-domain versus single-domain cognitive training in non-demented older people: a randomized controlled trial. BMC Med. 10, 30 (2012). • Depoortère, R., Auclair, A.L., Bardin, L., Colpaert, F.C., Vacher, B., Tancredi, A.N. F15599, a preferential post-synaptic 5-HT 1A receptor agonist: Activity in models of cognition in comparison with reference 5-HT 1A receptor agonists. Eur. Neuropsychopharmacol. 20, 641-654 (2010). • Diamond, M.C., Ingham, C.A., Johnson, R.E., Bennett, E.L. & Rosenzweig, M.R. (1976) Effects of environment on morphology of rat cerebral cortex and hippocampus . J. Neurobiol. 7, 75-85 (1976). • Erickson, C.A. & Barnes, C.A. The neurobiology of memory changes in normal aging. Exp Gerontol, 38, 61-69 (2003). • Erickson, K.I., Voss, M.W., Prakash, R.S., Basak, C., Szabo, A., Chaddock, L., Kim, J.S., Heo, S., Alves, H., White, S.M., Wojcicki, T.R., Mailey, E., Vieira, V.J., Martin, S.A., Pence, B.D., Woods, J.A., McAuley, E. & Kramer, A.F. Exercise training increases size of hippocampus and improves memory. Proc. Natl. Acad. Sci. USA. 108, 3017-3022 (2011). • Faherty, C.J., Kerley, D. & Smeyne, R.J.A. Golgi-Cox morphological analysis of neuronal changes induced by environmental enrichment. Brain Res. Dev. Brain Res. 144, 55-61 (2003). • Fischer, A. Environmental enrichment as a method to improve cognitive function. What can we learn from animal models? Neuroimage. 131, 42-7 (2016). • Freret, T., Billard, J. M., Schumann-Bard, P., Dutar, P., Dauphin, F., Boulouard, M. Bouet, V. Rescue of cognitive aging by long-lasting environmental enrichment exposure initiated before median lifespan. Neurobiol. Aging. 33, 1-10 (2012). • Frick, K.M. & Benoit, J.D. Use it or lose it: environmental enrichment as a means to promote successful cognitive aging. Scientific World Journal. 10, 1129-114 (2010). • Galeano, P., Blanco, E., Logica Tornatore, T.M., Romero, J.I., Holubiec, M.I., Rodríguez de Fonseca, F. & Capani, F. Life-long environmental enrichment counteracts spatial learning, reference and working memory deficits in middle-aged rats subjected to perinatal asphyxia. Front. Behav. Neurosci. 8, 406 (2015).
134 • Gibb, R. & Kolb B. A method for vibratome sectioning of Golgi-Cox stained whole rat brain. J. Neurosci. Methods. 79, 1-4 (1998). • Gould, E., Beylin, A., Tanapat, P., Reeves, A. & Shors, T.J. Learning enhances adult neurogenesis in the hippocampal formation. Nat. Neurosci. 2, 260-265 (1999). • Green, E.J., William, T., Greenough & Schlumpf, B.E. Effects of complex or isolated environments on cortical dendrites of middle-aged rats. Brain Res. 264, 233-240 (1983). • Greenough, W.T. & Volkmar, F.R. Pattern of dendritic branching in occipital cortex of rats reared in complex environments. Exp. Neurol. 40, 491-504 (1973). • Gundersen, H.J., Bendtsen, T.F., Korbo, L., Marcussen, N., Moller, A., Nielsen, K., Nyengaard, J.R., Pakkenberg, B., Sorensen, F.B., Vesterby, A. & West, M.J. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS . 96, 379-394 (1988). • Harburger, L.L., Nzerem, C.K., Frick, K.M. Single enrichment variables differentially reduce agerelated memory decline in female mice. Behav. Neurosci. 121, 679-688 (2007). • Hempel, A., Giesel, F.L., Garcia Caraballo, N.M., Amann, M., Meyer, H., Wüstenberg, T., Essig, M. & Schröder, J. Plasticity of cortical activation related to working memory during training. Am. J. Psychiatry .161, 745-747 (2004). • Jacobs, B., Schall, M. & Scheibel, A.B. A quantitative dendritic analysis of Wernicke’s area in humans. II. Gender, hemispheric, and environmental factors. J. Comp. Neurol. 327, 97-111 (1993). • Jiang, L., Cao, X., Li, T., Tang, Y., Li, W., Wang, J., Chan, R.C. & Li, C. Cortical thickness changes correlate with cognition changes after cognitive training: evidence from a chinese community study. Front. Aging Neurosci. 8, 118 (2016). • Juraska, J.M., Fitch, J.M. & Wasbume, D.I. The dendritic morphology of pyramidal neurons in the rat hippocampal CA3 area. II. Effects of gender and experience. Brain Res. 79, 115-121 (1989). • Juraska, J.M., Fitch, J.M., Henderson, C. & Rivers, N. Sex differences in the dendritic branching of dentate granule cells following differential experience. Brain Res. 333, 73-80 (1985). • Kempermann, G., Gast, D., & Gage, F.H. Neuroplasticity in old age: sustained fivefold induction of hippocampal neurogenesis by long-term environmental enrichment. Ann. Neurol. 52, 135-143 (2002). • Kesner, R.P. Subregional analysis of mnemonic functions of the prefrontal cortex in the rat. Psychobiology. 28, 219-228 (2000).
135 • Keuker, J.I., Vollmann-Honsdorf, G.K. & Fuchs, E. How to use the optical fractionator: an example based on the estimation of neurons in the hippocampal CA1 and CA3 regions of tree shrews. Brain Res. Brain Res. Protoc. 7, 211-221 (2001). • Kolb, B., Gorny, G., Söderpalm, A.H. & Robinson, T.E. Environmental complexity has different effects on the structure of neurons in the prefrontal cortex versus parietal cortex or nucleus accumbens. Synapse. 48, 149-153 (2003). • Maguire, E.A., Woollett, K. & Spiers, H.J. London taxi drivers and bus drivers: a structural MRI and neuropsychological analysis. Hippocampus. 16, 1091-1101 (2006). • Milgram, N.W., Siwak-Tapp, C.T., Araujo, J. & Head, E. Neuroprotective effects of cognitive enrichment. Ageing Res. Rev. 5, 354-369 (2006). • Mohammed, A. H., Zhu, S. W., Darmopil, S., Hjerling-Lefler, J., Ernfors, P., Winblad, B., Diamond, M.C., Eriksson, P.S. & Bogdanovic, N. Environmental enrichment and the brain. Prog Brain Res. 138, 109-133 (2002). • Mora, F., Segovia, G. & Del Arco, A. Aging, plasticity and environmental enrichment: structural changes and neurotransmitter dynamics in several areas of the brain. Brain Res. Rev. 55, 78-88 (2007). • Morris, R.G.M. Spatial localization does not require the presence of local cues. Learn. Mem. 12, 239260 (1981). • Morris, R. Developments of a water-maze procedure for studying spatial learning in the rat. J. Neurosci. Methods. 11, 47-60 (1984). • Mota, C., Pinto, V., Sousa, N., Sousa, J.C. & Cerqueira, J.J. Bigger is worse: volumetric correlates of age-related changes in hippocampus and medial prefrontal cortex in the rat. Under preparation. • Mota, C., Taipa, R., Pereira das Neves, S., Monteiro-Martins, S., Monteiro, S., Palha, J.A., Sousa, N., Sousa, J.C. & Cerqueira, J.J. Structural and molecular correlates of cognitive aging in the rat. Manuscript accepted for publication in Scientific Reports, 2018. • Nouchi, R., Taki, Y., Takeuchi, H., Hashizume, H., Akitsuki, Y., Shigemune, Y., Sekiguchi, A., et al., Brain training game improves executive functions and processing speed in the elderly: a randomized controlled trial. PloS One. 7, 29676; 10.1371/journal.pone.0029676 (2012.) • Olesen, P.J., Westerberg, H. & Klingberg, T. Increased prefrontal and parietal activity after training of working memory. Nat. Neurosci. 7, 75-79 (2004). • Palomero-Gallagher, N. & Zilles, K. Isocortex in The Rat Nervous System (ed. Paxinos, G.) 729-757 (Academic, 2004).
136 • Paxinos, G. & Watson, C. The Rat Brain In Stereotaxic Coordinates (ed. Paxinos, G.) (Academic, 1998). • Peinado, M.A., Quesada, A., Pedrosa, J.A., Martinez, M., Esteban, F.J., Del Moral, M.L. & Peinado, J.M. Light microscopic quantification of morphological changes during aging in neurons and glia of the rat parietal cortex. Anat. Rec. 247, 420-425 (1997). • Pinto, V., Costa, J.C., Morgado, P., Mota, C., Miranda, A., Bravo, F.V., Oliveira, T.G., Cerqueira, J.J. & Sousa, N. Differential impact of chronic stress along the hippocampal dorsal-ventral axis. Brain Struct. Funct. 220, 1205-1212 (2015). • Rampon, C., Jiang, C.H., Dong, H., Tang, Y.P., Lockhart, D.J. Schultz, P.G., Tsien, J.Z. & Hu, Y. Effects of environmental enrichment on gene expression in the brain. Proc. Natl. Acad. Sci. USA. 97, 12880-12884 (2000). • Rebok, G.W., Ball, K., Guey, L.T., Jones, R.N., Kim, H.Y., King, J.W., et al. Ten-year effects of the advanced cognitive training for independent and vital elderly cognitive training trial on cognition and everyday functioning in older adults. J. Am. Geriatr. Soc. 62, 16-24 (2014). • Richards, M., Hardy, R. & Wadsworth, M.E. Long-term effects of breast-feeding in a national birth cohort:educational attainment and midlife cognitive function. Public Health Nutr. 5, 631-635 (2002). • Santos, N.C., Costa, P.S., Cunha, P., Cotter, J., Sampaio, A., Zihl, J., Almeida, O.F.X., Cerqueira, J.J., Palha, J.A. & Sousa, N. Mood is a key determinant of cognitive performance in communitydwelling older adults: a cross-sectional analysis. Age. 35, 1983-1993 (2013). • Seider, T.R., Fieo, R.A., O’Shea, A., Porges, E.C., Woods, A.J. & Cohen, R.A. Cognitively engaging activity is associated with greater cortical and subcortical volumes. Front. Aging Neurosci. 8, 1-10 (2016). • Shors, T.J., Anderson, M.L., Curlik 2nd, D.M. & Nokia, M.S. Use it or lose it: how neurogenesis keeps the brain fit for learning. Behav. Brain Res. 227, 450-458 (2012). • Singh-Manoux, A., Kivimaki, M., Glymour, M.M., Elbaz, A., Berr, C., Ebmeier, K.P., Ferrie, J.E. & Dugravot, A. Timing of onset of cognitive decline: results from Whitehall II prospective cohort study. BMJ. 344, 7622; 10.1136/bmj.d7622 ( 2012). • Sousa, N. The dynamics of stress neuromatrix. Mol. Psychiatry. 21, 302-312 (2016). • Stein, L.R., O’Dell, K.A., Funatsu, M., Zorumski, C.F. & Izumi, Y. Short-term environmental enrichment enhances synaptic plasticity in hippocampal slices from aged rats. Neuroscience. 329, 294-.305 (2016).
137 • Steiner, B., Zurborg, S., Hörster, H., Fabel, K. & Kempermann, G. Differential 24h responsiveness of Prox1-expressing precursor cells in adult hippocampal neurogenesis to physical activity, environmental enrichment, and kainic acid-induced seizures. Neuroscience . 154, 521-529 (2008). • Stern, Y. Cognitive reserve in ageing and Alzheimer's disease. Lancet Neurol. 11, 1006-1012 (2012). • Tanigawa, T., Takechi, H., Arai, H., Yamada, M., Nishiguchi, S. & Aoyama, T. Effect of physical activity on memory function in older adults with mild Alzheimer’s disease and mild cognitive impairment. Geriatr. Gerontol. Int. 14, 758-762 (2014). • Uylings, H.B., Ruiz-Marcos, A. & van Pelt, J. The metric analysis of three-dimensional dendritic tree patterns: a methodological review. J. Neurosci. Methods. 18, 127-151 (1986). • Uylings, H.B. & van Pelt, J. Measures for quantifying dendritic arborizations. Network. 13, 397-414 (2002). • Van der Staay, F.J. Spatial working memory and reference memory of Brown Norway and WAG rats in a holeboard discrimination task . Neurobiol. Learn. Mem. 71, 113-125 (1999). • Van der Staay, F.J., Gieling, E.T., Pinzón, N.E., Nordquist, R.E. & Ohl, F. The appetitively motivated "cognitive" holeboard: a family of complex spatial discrimination tasks for assessing learning and memory. Neurosci Biobehav. Rev. 36, 379-403 (2012). • van Praag, H., Kempermann G. & Gage F.H. Running increases cell proliferation and neurogenesis in the adult mouse dentate gyrus. Nat. Neurosci. 2, 266-270 (1999). • Vicens, P., Bernal, M.C., Carrasco, M.C. & Redolat, R. Previous training in the water maze: differential effects in NMRI and C57BL mice. Physiol. Beh. 67, 197-203 (1999). • Vicens, P., Redolat, R. & Carrasco, M.C. Effects of early spatial training on water maze performance: a longitudinal study of mice. Exp. Gerontol . 37, 575-581 (2002). • Vogt, B.A., Vogt, L. & Farber, N. Cingulate cortex and disease models in The Rat Nervous System (ed. Paxinos, G.) 705-727 (Academic, 2004). • Vorhees, C.V. & Williams, M.T. Assessing Spatial Learning and Memory in Rodents. ILAR Journal. 55, 310-332 (2014). • West, M.J., Slomianka, L. & Gundersen, H.J. Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator. Anat. Rec. 231, 482-497 (1991). • Whalley, L.J., Deary, I.J., Appleton, C.L. & Starr, J.M. Cognitive reserve and the neurobiology of cognitive aging. Ageing Res. Rev. 3, 369-382 (2004).
138 • Wilson, R.S., Bennett, D.A., Bienias, J.L., Aggarwal, N.T., Mendes de Leon, C.F., Morris, M.C., Schneider, J.A. & Evans, D.A. Cognitive activity and incident AD in a population based sample of older persons. Neurology. 59, 1910-1914 (2002). • Wurm, F., Keiner, S., Kunze, A., Witte, O.W. & Redecker, C. Effects of skilled forelimb training on hippocampal neurogenesis and spatial learning after focal cortical infarcts in the adult rat brain. Stroke. 38, 2833-2840 (2007).
139 Table 1| Results of two-way ANOVA on the data from Pseudo-HB and HB animals. Training (HB VS P-HB) Reference memory performance Interaction df F P ηρ² F P ηρ² F P ηρ² Hippocampal formation Morphology of neurons Granular neurons 1,24 1.730 0.201 0.067 2.893 0.102 0.108 7.751 0.010 0.244 CA3 pyramidal neurons (apical dendrite) 1,24 0.822 0.374 0.033 0.022 0.882 0.001 0.095 0.760 0.004 CA3 pyramidal neurons (basal dendrites) 1,24 0.037 0.849 0.002 3.276 0.083 0.120 0.020 0.889 0.001 CA1 pyramidal neurons (apical dendrite) 1,24 2.132 0.157 0.082 2.410 0.134 0.091 1.942 0.176 0.075 CA1 pyramidal neurons (basal dendrites) 1,24 1.288 0.268 0.051 0.886 0.356 0.036 0.006 0.939 0.000 Sholl analysis Granular neurons 1,24 0.845 0.367 0.034 3.132 0.089 0.115 6.839 0.015 0.222 CA3 pyramidal neurons (apical dendrite) 1,24 0.782 0.385 0.032 0.001 0.972 0.000 0.007 0.936 0.000 CA1 pyramidal neurons (apical dendrite) 1,24 3.100 0.091 0.114 1.943 0.176 0.075 0.992 0.329 0.040 Volumetric measurements Dorsal HPC 1,19 3.367 0.082 0.151 0.000 0.999 0.000 8.291 0.010 0.304 Dorsal – DG molecular layer 1,19 3.769 0.067 0.166 0.125 0.727 0.007 3.867 0.064 0.169 Dorsal – DG granular layer 1,19 0.010 0.920 0.001 0.130 0.723 0.007 5.680 0.028 0.230 Dorsal – DG hilus 1,19 25.639 <0.0005 0.574 2.563 0.126 0.119 4.168 0.055 0.180 Dorsal – CA3 stratum oriens 1,19 0.049 0.827 0.003 0.128 0.724 0.007 2.828 0.109 0.130 Dorsal – CA3 pyramidal layer 1,19 0.052 0.822 0.003 0.307 0.586 0.016 0.836 0.372 0.042 Dorsal – CA3 stratum radiatum 1,19 1.192 0.289 0.059 0.616 0.442 0.031 0.279 0.603 0.014 Dorsal – CA1 stratum oriens 1,20 0.683 0.418 0.033 0.021 0.886 0.001 0.234 0.634 0.012 Dorsal – CA1 pyramidal layer 1,20 8.435 0.009 0.297 0.125 0.728 0.006 0.245 0.626 0.012 Dorsal – CA1 stratum radiatum 1,20 0.465 0.503 0.023 0.016 0.899 0.001 0.653 0.428 0.032 Total number of neurons Dorsal DG granular layer 1,19 63.035 <0.0005 0.768 7.528 0.013 0.284 1.512 0.234 0.074 CA3 stratum pyramidal 1,20 10.864 0.004 0.352 0.489 0.492 0.024 0.591 0.451 0.029 CA1 stratum pyramidal 1,20 40.583 <0.0005 0.670 0.050 0.825 0.002 2.989 0.099 0.130