*4.3. Different Behavioral Signatures for Physical, Emotional, and Cognitive Phenotypes*

The patterns of innate neophobia (fear of novelty) response shown by NTg mice were found broken in the 3xTg-AD genotype, and it is most prominently observable in the mutants with female sex, as consistently described since our first work [10] and confirmed afterwards [21,37,49]. We proposed increased neophobia, as delayed and reduced rearing, an early behavioral marker of the onset of behavioral and psychological symptoms of dementia (BPSD)-like symptoms since premorbid disease stages [28]. Here, the corner test was sensitive to old 3xTg-AD females' genotype, where neophobia is enhanced. For the first time, we describe these patterns as also observable on the repeated test.

In the open-field test, no sex differences were found in normal aging. In contrast, the frail male 3xTg-AD exhibited sustained activity, mostly as a thigmotaxis response and slower habituation pattern. A hyperactive pattern in frail 3xTg-AD males is also observed after social isolation [50]. The repetition of the test elicited reduced activity, in all the groups. In agreement with previous reports showing a 24 h long-term memory deficit in male 3xTg-AD mice at 2, 4, and 6 months of age [28], the behavioral response did not benefit from previous experience. Here, we show that these genotype effects in the immediate re-confrontation with the test are extensive to 14 months of age and mostly observed in females.

As shown by increased crossings and time in the lit area and grooming, disinhibitory behavior in the dark–light box confirmed a consistent BPSD-like phenotype in male and female 3xTg-AD mice. These disinhibitory patterns were first described at 4 months of age as part of the profile mimicking AD's prodromal stage [51]. The marble test, assessing anxiety-like behaviors and screen drugs for obsessive-compulsive disorders and psychotic symptoms [52], also showed a specific pattern for females 3xTg-AD mice.

In the paradigms for learning and memory in mazes [37], the number of animals failing to complete the T-maze indicates their aged status and/or poor motivation [8]. The latency to achieve the first goal of the test (crossing the intersection) has been related to immunosenescence and reduced survival [53]. This ceiling effect resulted in a sample of "successful animals" that equally performed the spontaneous alternation task. However, these animals committed errors attributed to working memory (revisiting explored areas) when assessed in a more complex task (the forced-choice paradigm) in a sex-specific manner. Here, females of both genotypes spent more time choosing the right choice and committed more errors.

The water maze's performance was strongly determined by genotype differences in the swimming speed (slower in 3xTg-AD mice) mostly found among females. Motor features can be discarded (the frail animals were males), but the swimming performance can reflect their emotional status in an aquatic environment known to be anxiogenic for mice [54]. To control this factor, the distance covered was used instead of the latency. Two learning and memory tasks differing on the level of complexity and involving short (15 min) and long-term (24 h) memory were used. The day-by-day and trial-by-trial analysis showed a notorious aged profile compared to previous reports in young [10,11,28,37,49,51,55] or old animals [8,38,48,56]. Worse performance of female NTg mice was observed on the visual

perceptual learning task. Long-term spatial reference learning and memory deficits were shown by all the groups in their first day of place learning task where the hidden platform, located in a reversed location, had to be found. Finally, the two-day place task indicated paradoxical better performances in male 3xTg-AD mice, which could be explained by the strong need that a frail animal may have to find and remember a safe place.

#### *4.4. Increased MRI-ASL Regional Cerebral Blood Flow in 3xTg-AD Survivor Females*

In the present study, we evaluated CBF in five different brain regions, namely, hippocampus, cortex, striatum, caudate putamen, and amygdala using arterial spin labeling (ASL), a magnetic resonance imaging (MRI) technique for non-invasive measurements of cerebral blood blow. The results indicated sex- and brain-region-associated changes in CBF. Among all, 3xTg-AD female mice survivors had increased CBF in the cortex and hippocampus as compared with their wild-type counterparts.

Although CBF alteration seems to be involved in AD pathogenesis, the perfusion patterns remain unclear, since both hypoperfusion and hyperperfusion have been described in different brain areas and involved in different brain functions [57].

Oxidative stress, inflammation, and cerebrovascular disease have been suggested to be involved in AD. In one of our recent collaborative studies in 3xTg-AD mice, we have reported that the number of β-amyloid (Aβ) plaques in the hippocampus and entorhinal cortex at advanced stages of the disease was higher in females than in males [58]. Interestingly, co-localization of hypoxic areas and Aβ plaques in the hippocampus and entorhinal cortex were observed only in females. In the present study, the increased CBF in the cortex and hippocampus in female survivors suggests a potential compensatory hemodynamic mechanism in end-of-life dementia, which is sex- and brain-region-dependent. This is interesting to note, since recent work in APP/PS1 transgenic mice at the early stages of AD has shown longitudinal changes in regional CBF, indicating age- and brain-regiondependent alterations of cerebral blood flow [59]. At the clinical level, an increased CBF has also been observed in patients at preclinical stages of AD when cognitive performance is still preserved, suggesting a compensatory response to the accumulation of Aβ pathology [60]. Moreover, the reduction in CBF seems to be an important factor contributing to the cognitive dysfunction associated with dementia. One study performed with APP/PS1 mice in the late stages of the disease reported that a treatment that consists of the increase in the cerebral blood flow improves cognition [61].
