The present thesis explored a number of topics related to pathogenic and pathological aspects of Alzheimer's disease (AD), the most common type of age-related dementia. This devastating disease is characterized by progressive neuronal atrophy, cognitive decline, deterioration of language, personality and psychiatric abnormalities. There are currently ~35.6 million elderly individuals afflicted with AD worldwide and epidemiological studies predict this figure to quadruple by 2050, pushing the global economic burden of AD to extremes. Only limited symptomatic treatments are currently available for AD, and this calls for novel therapeutic approaches to combat the disease. Yet recent reviews highlight the worrying decline in the number of clinical trials for AD in the last decade. The failure rate of ~99% in clinical AD trials (compared to cancer ~81%) suggests that the current biological markers may not represent the pathogenic culprits of AD and yet novel therapeutic targets need to be identified and validated. Pathological end-stage hallmarks of AD include the appearance of extracellular amyloid plaques composed of the amyloid ß (Aß) peptide and intracellular neurofibrillary tangles (NFT) aggregated from paired helical fillaments of hyperphosphorylated tau protein. Gene mutations associated with abnormal processing of Aß-precursor protein (APP) and tau have been identified in pedigrees affected by familial forms of dementia. However, less than 1% of AD population suffers from the early onset familial AD (fAD) caused by the inheritance of mutations within APP or presenilin (PS) genes, and the vast majority of cases are sporadic (sAD). Amongst the multiple risk factors for sAD are old age, poor diet and a medical history of metabolic disorders, such as diabetes. Interestingly, despite divergent aetiology between fAD and sAD, both conditions are characterized by similar morphological brain changes, Aß, NFT pathology, neuronal energy decline and neurodegeneration. Therefore, fAD-related gene mutations continue to aid preclinical AD research and provide fundaments for experimental modelling of the disease. Genetic manipulation in animals offers the only path to understand the complex nature of AD and may effectively lead to a development of preventive strategies. Although numerous mouse models genetically designed to develop AD-like phenotypes have been engineered, no mouse model has yet fully recapitulated the age-dependent and progressive pathophysiology of human AD. The majority of existing AD-like mice have so far relied on a gross over-expression of fAD gene mutations, whereby multiple copies of human mutant transgenes are inserted via pronuclear injection. However, there are a number of drawbacks associated with the pathology in these mice, such as genetic stability, anomalies induced by the gene insertion and the lack of prodromal stage. The present thesis focused on a characterization of a new generation of mouse lines, termed PLB, created with the aim of modelling AD without gross mutant protein over-expression. PLB1/2 mice were generated via targeted knock-in (KI) of human AD-related genes (hAPP/hTau) under the control of CaMKIIa (calcium/calmodulin-dependent kinase IIa) promoter. The resultant PLB1Double transgenic line harbouring a double mutant gene construct (Swedish and London mutation within the APP gene; P301L and R306W within the tau gene) was subsequently crossed with the pre-existing hPS1 mouse (A246E mutation) to obtain the polygenic PLB1Triple mouse (Platt et al., 2011). The genetic design of PLB1Double mouse permitted the generation of single transgenic mice (PLB2APP and PLB2Tau) via selective deletion of either hAPP or hTau cDNA floxed between specific flanking sites. Additionally, the single transgenic PLB2APP line was crossed with the hPS1 mouse to generate a double transgenic PLB2APP/PS1 model, allowing for multiple inter-strain comparisons between single, double and triple transgenic mice. Here, PLB1/2 mouse lines were characterized at 6 and 12 months with a focus on ADand age-related behavioural and molecular phenotypes. Cognitive testing was performed using the standard water maze (WM) and a novel semantic-like memory test, termed social transmission of food preference (STFP). Activity and anxiety-related phenotypes were also investigated in transgenic PLB lines. Neuronal tissue was subsequently harvested from selected PLB animals and screened for inflammation, Aß, and tau pathology. Briefly, all hAPP-containing PLB1/2 mice were impaired in spatial acquisition in the WM from 6 months, and deficits were most prominent in aged PLB2APP/PS1 and PLB1Triple mice. Interestingly, unlike other transgenics, PLB2Tau mice did not present with spatial deficits in WM, but exhibited a robust impairment in the semantic-like memory test. Circadian rhythms were preserved in all PLB lines, but drastic reductions in locomotion occurred during both light (range: -21% to - 38%) and dark periods (-22% to -33%). A graded robustness of hypoactive phenotypes were seen across PLB lines, with single transgenics (PLB2APP and PLB2TAU) affected less cf. controls, while PLB1Triple mice exhibited the most profound deficit. Interestingly, only the hTau-expressing mice (PLB2Tau and PLB1Triple) displayed heightened anxiety-like behaviours in the elevated-plus maze (EPM). These results overall suggested that while hAPP expression induced deficits in hippocampus-dependent tasks (spatial WM), hTau promoted increased fear responses and cognitive deficits associated with cortical function (semantic STFP).