International Journal of Molecular Sciences Article Sex- and Neuropsychiatric-Dependent Circadian Alterations in Daily Voluntary Physical Activity Engagement and Patterns in Aged 3xTg-AD Mice Daniel Alveal-Mellado 1,2 , Lidia Castillo-Mariqueo 1,2 and Lydia Giménez-Llort 1,2, * 1 Institut de Neurociències, Universitat Autònoma de Barcelona, 08193 Barcelona, Spain 2 Department of Psychiatry and Forensic Medicine, School of Medicine, Universitat Autònoma de Barcelona, 08193 Barcelona, Spain * Correspondence: lidia.gimenez@uab.cat; Tel.: +34-93-5812378 Abstract: Alzheimer’s disease (AD) patients suffer from circadian rhythm alterations affecting their daily physical activity patterns with less willingness to perform a voluntary exercise. In preclinical studies, there is no clarity on whether animal models of AD can replicate these impairments. Here, we provide a proof of concept of the performance and behavioral effects of four weeks of voluntary wheel running (VWR) in a group of 14-month-old male and female 3xTg-AD mice at advanced stages of AD and the daily variance (behavioral circadian rhythmicity) of VWR associated with sex and their neuropsychiatric-like phenotype. Higher levels of horizontal exploration in the open field (OF) test were found in mice submitted to exercise. A linear mixed effect model showed significant sex-dependent differences in the VWR activity performed on the first night of follow-up, with high- NIBI males running less than high-NIBI females. Thus, an influence of NPS-like symptoms on the circadian patterns of VWR may account for such differences. In addition, males remained more active Citation: Alveal-Mellado, D.; than females during diurnal periods. We hypothesize that this increment in energy expenditure Castillo-Mariqueo, L.; Giménez-Llort, during resting periods may be related to hyperactive behavior, similar to that observed in humans’ L. Sex- and Neuropsychiatric- exacerbated agitation or sundowning behavior. These findings support the usage of the 3xTg-AD Dependent Circadian Alterations in mouse as a reliable model for studying circadian rhythm alterations in AD and, at the translational Daily Voluntary Physical Activity level, the importance of tailored and individualized physical activity programs in clinical settings. Engagement and Patterns in Aged 3xTg-AD Mice. Int. J. Mol. Sci. 2022, Keywords: Alzheimer’s disease; animal model; sex differences; rehabilitation; exercise; physical 23, 13671. https://doi.org/10.3390/ activity; circadian rhythms ijms232213671 Academic Editor: Fuyuki Sato Received: 2 October 2022 1. Introduction Accepted: 3 November 2022 Published: 8 November 2022 An insidious cognitive decline comprising anterograde memory and executive func- tion impairments are the main diagnostic criteria for dementia caused by Alzheimer’s Publisher’s Note: MDPI stays neutral disease (AD) [1]. In addition, a broad spectrum of neuropsychiatric symptoms (NPS) may with regard to jurisdictional claims in accompany the instauration of the disease [2], including affective disorders (anxiety and published maps and institutional affil- depression), behavioral and emotional difficulties (apathy and mood fluctuation), psychotic iations. symptoms (hallucinations and delusions), and sundowning behavior (circadian rhythm alterations) [3]. Indeed, when severe, such dysfunctions represent the leading cause of institutionalization [4,5]. Because of the heterogeneity of the clinical phenomena, NPS are Copyright: © 2022 by the authors. also referred to as behavioral and psychological symptoms of dementia (BPSD) [6,7]. Licensee MDPI, Basel, Switzerland. Circadian rhythm dysfunctions (CRD) are also reported in the early stages of the This article is an open access article disease, and it comprises sleep, thermoregulation, and movement disorders [8]. CRD are distributed under the terms and also observed in the healthy aging process due to changes in afferent pathways to the conditions of the Creative Commons circadian pacemaker of the organism, the suprachiasmatic nucleus (SNC) [9]. However, AD Attribution (CC BY) license (https:// patients suffer more pronounced changes, and degeneration affects the SCN even before creativecommons.org/licenses/by/ the cognitive decline begins [10]. Furthermore, CRD have been proposed as a preclinical 4.0/). sign of the disease [11]. Therefore, nonpharmacological strategies that can be implemented Int. J. Mol. Sci. 2022, 23, 13671. https://doi.org/10.3390/ijms232213671 https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2022, 23, 13671 2 of 15 in individuals at risk or be part of the individuals’ lifestyle can play an important role as prevention strategies during asymptomatic and prodromal stages to hamper or counteract the underlying detrimental processes. Environmental factors such as maintaining moderate levels of physical activity (PA) may prevent cognitive decline [12,13] and modify neuropathological changes that occur in the early stages of AD [14]. However, clinical evidence warns that NPS/BPSD and CRD negatively influence routine exercise and physical rehabilitation engagement in these patients [15,16]. Thus, low levels of PA have been reported in free-living and institutional- ized AD patients, with a marked reduction in women [16–18]. At the translational level, the effects of PA have also been studied in animal models of AD. Thus, improvements in cognitive and BPSD-like symptoms are reported at the early and moderate stages of the disease [19,20], whereas the effects of PA at more advanced stages remain unclear. We have also shown that, despite the benefits, a forced exercise paradigm may also exert some minor adverse effects on females [21]. In the present work, we addressed the animal’s engagement in the voluntary wheel-running paradigm for the first time. Equally important have been the study of CRD in aging and AD using rodent models. Daily movement activity variations, measured as “in cage” activity or running wheel (RW) activity, are frequently reported with a general age-dependent decline [21] and greater levels of PA in females [22,23], probably attributable to a protector effect of estrous cycle hormones [24]. The triple transgenic model for AD (3xTg-AD) developed by Frank LaFerla’s labo- ratory [25] has shown good face and construct validity [26], replicating the extracellular beta-amyloid plaques and intracellular neurofibrillary tangles accumulation within specific brain areas, such as the hippocampus and neocortex, considered the neuropathological hallmarks of the disease [25,27,28]. Cognitive symptoms arise in animals’ early ages and can be detected in tests evaluating spatial orientation and working memory (i.e., Morris water maze and T-maze) [29]. On the other side, tests involving novel environments (i.e., corner test and open-field test) have shown sensible to detect noncognitive symptoms (NPS-like symptoms) through the induction of higher freezing behavior and reduced horizontal exploration. These findings have been previously referred to as novelty-induced behavioral inhibition (NIBI) [30,31]. Regarding CRD, the model exhibits circadian alterations [29] and reproduces sleep and movement activity disorders at 6 months old [32,33]. Nevertheless, the temporal progression of such impairments at older ages has not been fully depicted. The present work aims were (1) to model the pattern of behavioral circadian rhythmic- ity of PA in a group of 3xTg-AD animals at advanced ages submitted to daily voluntary wheel running (VWR) and (2) to evaluate the effects of 30 days of PA in noncognitive symptoms of AD. 2. Results Body weight was monitored from the day before the start of the behavioral battery of tests until the week after the end of the 30-day follow-up (see Figure 1). Weight variation differences were analyzed through a 2 × 2 × 2 (sex × treatment × days) ANOVA test showing a treatment effect [T, F(1,30) = 11.325, p = 0.002] with a significant decrease in weight after 30 days in the VWR group [SED vs VWR, t(18,16) = 2.97, p = 0.006, t-test], mostly due to statistically significant differences observed in males [SED males vs VWR males, t(10,7) = 4.16, p = 0.001, t-test]. Indeed, the sex × treatment interaction effect [F(1,30) = 5.70, p = 0.023] shows that SED females also showeda negative weight variation. Int. Int.J.J.J.Mol. Int. Mol.Sci. Mol. Sci. 2022,23, Sci.2022, 2022, 23, 13671 23, 13671 13671 33 of 16 3 of 15 of 16 Figure 1. Body weight (A) at baseline (week 1) and after 30 days follow-up (week 6). The intervention Figure 1. Body weight (A) at baseline (week 1) and after 30 days follow‐up (week 6). The interven‐ tooktook tion place between place between weeks weeks22andand 5. (B)Weight 5. (B) Weightvariation variationperper group group (weight (weight at week at week 6 -week 6 ‐week 1). 1). Factorialanalysis: Factorial analysis:D,D, days; days; T, T, treatment; treatment; S, sex. S, sex. p < 0.05; * p <* 0.05; *** p*** p < 0.001. < 0.001. 2.1. Corner (CT) and Open Field (OF) Tests 2.1. Corner (CT) and Open Field (OF) Tests Inthe In theCT, CT,no nobaseline baseline differences differences between between groups groups werewere found found (Figure (Figure 2). By2). By contrast, contrast, aa 22××2 2× × 2 (sex 2 (sex × treatment × treatment × days) × days) mixedmixed (split-plot) (split‐plot) ANOVA ANOVA test yielded test yielded a days a days effect in effect in almost almost all variables all variables evaluated evaluated (D, all(D, all Fs(1,30) Fs(1,30) > 14.09, > 14.09, p < 0.001), p < 0.001), indicating indicating a reduction a reduction in in exploratory exploratory activity activity and anand an increment increment in neophobia in neophobia in the pre-post in the pre‐post analysis.analysis. Thus, all Thus, an‐ all animals imals (n =(n34) = 34) reduced reduced the number the number of corners of corners at 30ats (t(34) 30 s (t(34) = 6.35,= 6.35, p < 0.001, p < 0.001, t‐test) t-test) pairedpaired and60 and 60ss(t(34) (t(34)= = 6.51, 6.51, p 56.6, pared with SED animals [VWR vs. SED, all t(16,18) > 56.6, p < 0.015, t‐test]. Regarding p < 0.015, t-test]. Regarding verticalactivity, vertical sex××treatment activity,aasex treatment × days × days interaction interaction waswas found found T ××D, [S × [S × D, F(1,30) T F(1,30) = 5.07,= 5.07, pp==0.032, 0.032,Mixed Mixed ANOVA]. ANOVA]. A further A further analysis analysis showed showed thatthat significance significance couldcould be explained be explained bysex by sexdifferences differencesininthe the VWR VWR group group at baseline at baseline [3xTg-AD [3xTg‐AD malemaleVWRVWR vs. 3xTg-AD vs. 3xTg‐AD femalefemale VWR, t(7,9) VWR, 2.4,pp==0.042]. t(7,9)==2.4, 0.042].NoNodifferences differenceswerewerefound foundininthethelatency latencyofofrearing, rearing,thethela‐latency tency of grooming, of grooming, or emotionality or emotionality (urination (urination andand defecationboli) defecation boli)(data (data not not shown). shown).Fi‐ Finally, nally, whenwhen the distance the distance covered covered by by thethe animals animals wasanalyzed, was analyzed, aa daysdays effect effectwas wasfound foundinin the the central central zone zone [D,[D, F(1,30) F(1,30) = 12.60, = 12.60, p =p 0.001] = 0.001] with with a decreaseininallallgroups a decrease groupswhenwhenretested retested[retest [retest vs. baseline, vs. baseline, t(34) t(34) = 3.66,= 3.66, p < 0.001, p < 0.001, paired paired t‐testMoreover, t-test]. ]. Moreover, sexsex effectswere effects werefound foundin the in the periphery [S, F(1,30) = 7.98, p = 0.008] and total field [S, F(1,30) = periphery [S, F(1,30) = 7.98, p = 0.008] and total field [S, F(1,30) = 6.02, p = 0.02] after 6.02, p = 0.02] after 30 days 30 days with males covering with males covering more distance. more distance. Openfield Figure3.3.Open Figure fieldtest. test.(A) (A)Horizontal Horizontal activity activity latencies: latencies: movement movement latency latency (movement), (movement), latency to latency start exploring in the central zone (Zcentral), latency to reach the periphery (Zperiphery). to start exploring in the central zone (Zcentral), latency to reach the periphery (Zperiphery). (B) and (B) and (C).Vertical (C). Verticalactivity: activity:rearing rearing latency latency (LatR), (LatR), grooming grooming latency latency (LatG), (LatG), total total rearing rearing episodes episodes in 5 min in 5 min (TotalR), (TotalR),total totalgrooming grooming episodes episodes in in 5 min 5 min(TotalG) (D) (D) (TotalG) distance covered distance in theinOF covered theinOF theincentral the central zone (Zcentral), peripheral zone (Zperiphery), and total field (Ztotal). Factorial analysis: D, days; T, treatment; S, sex; s < 0.05 vs. male counterpart in the same treatment group. * p < 0.05; ** p < 0.01. 2.2. T-Maze Spontaneous Alternation (TMSA) The TMSA (Figure 4 test showed no differences in the latencies related to freez- ing behavior (to move and turn) and exploration; namely, the time until start to ex- zone (Zcentral), peripheral zone (Zperiphery), and total field (Ztotal). Factorial analysis: D, days; T, treatment; S, sex; s < 0.05 vs. male counterpart in the same treatment group. * p < 0.05; ** p < 0.01. Int. J. Mol. Sci. 2022, 23, 13671 5 of 15 2.2. T‐Maze Spontaneous Alternation (TMSA) The TMSA (Figure 4 test showed no differences in the latencies related to freezing behavior (to move and turn) and exploration; namely, the time until start to explore (Ex‐ plore plore),(Explore), reach the intersection reach the intersection (LatT), and the(LatT), and the time spent time until spentthe reaching until firstreaching (Arm1) and the first (Arm1) second (Arm2) arm of the maze, and the test completion criteria (End) [All Fs(1,29) < 3.49, [All and second (Arm2) arm of the maze, and the test completion criteria (End) Fs(1,29) p > 0.05, mixed p > 0.05, However, < 3.49,ANOVA]. mixed ANOVA]. However, a sex × treatment a sex × days × treatment interaction × days was found interac- in the tion time elapsed until crossing the intersection with their four paws (LatT4) [F(1,29) = 4.80, ppaws was found in the time elapsed until crossing the intersection with their four (LatT4) = 0.037, [F(1,29) = 4.80, pafter mixed ANOVA] = 0.037, mixed 30 days, ANOVA] with 3xTg‐AD after 30 days, males VWR with 3xTg-AD reaching males this goal in aVWR reaching this goal in a shorter time compared with 3xTg-AD females shorter time compared with 3xTg‐AD females VWR [t(7,8) = −2.20, p = 0.048, t‐test]. More‐ VWR [t(7,8) = −2.20, pover, = 0.048, t-test]. Moreover, considering the goals in considering the TMSAthe goals inasthe ethogram TMSA (pooled a whole ethogram as a awhole data), trend(pooled is data), a trend observable is better with observable with better performance performance of the 3xTg‐AD males of the 3xTg-AD VWR males compared withVWR compared all other with groupsall[Sother × T ×groups [S × D, F(1,29) T × pD,= F(1,29) = 3.90, 0.058] in=the 3.90, p = 0.058] retest. Instead,inno thesignificant retest. Instead, no signif- differences were found icant after 30were differences days found in the number after 30ofdays errorsinand thevertical number activity (rearings of errors and and groom‐ vertical activity ing) in theand (rearings (data not shown). test grooming) in the test (data not shown). Figure 4. TMSA test at baseline (left) and after 30 days follow-up (right). Latency to move (Move), Figure 4. TMSA test at baseline (left) and after 30 days follow‐up (right). Latency to move (Move), latency to turn (Turn), latency to reach the maze intersection (LatT) and cross the intersection with latency to turn (Turn), latency to reach the maze intersection (LatT) and cross the intersection with the four paws the four paws(LatT4), (LatT4),latency latencytoto explore explore thethe first first (Arm1) (Arm1) andand second second (Arm2) (Arm2) armsarms and latency and latency to to complete the thetest testcompletion completioncriteria criteria (End). (End). Factorial Factorial analysis: analysis: D, days; D, days; T, treatment; T, treatment; S, *sex. S, sex. * p < 0.05. p < 0.05. 2.3. Sensorimotor Assessment 2.3. Sensorimotor Assessment The sensorimotor The sensorimotorassessment assessmentcomprised comprised(1) (1)the themuscular muscularstrength strength(5(5s)s)andand muscular en- muscular durance endurance (60 s) grip tests evaluating forelimbs strength and (2) geotaxis, focusedthe (60 s) grip tests evaluating forelimbs strength and (2) geotaxis, focused on on response the to gravity assessment (Figure 5). In the muscular strength trial, a response to gravity assessment (Figure 5). In the muscular strength trial, a days effect [D, days effect [D, F(1,29) = 19.95, pF(1,29) < 0.001], sex × day [S × D, F(1,29) = 5.23, p = 0.03] and a sex × treatment = 19.95, p < 0.001], sex × day [S×D, F(1,29) = 5.23, p = 0.03] and a sex × treatment × × day interaction effects [S × T ×effects day interaction D, F(1,29) [S × T=×5.57, p = 0.025] D, F(1,29) were = 5.57, p = found. 0.025] wereA post hoc A found. analysis post hoc showed analysis that all groups showedincreased that all groupstheirincreased latency aftertheir30 days [t(34) latency after 30 =− 4.30, days p < =0.001, [t(34) −4.30,paired t-test], p < 0.001, pairedfemales last longer t‐test], thanlast females males afterthan longer 30 daysmales(3.21 vs.30 after 2.77 dayss) [t(17,17) (3.21 vs. =2.77−2.66, p = 0.01], s) [t(17,17) and 3xTg-AD = −2.66, p= VWR 0.01], andfemales had better 3xTg‐AD VWRperformance females had betterafter 30 days compared performance after with 30 daystheir values atwith compared baseline [t(9) −7.79,atpbaseline their =values < 0.001,[t(9) paired t-test]p vs. = −7.79, 3xTg-AD < 0.001, pairedSED females t‐test] [t(8,9) =SED vs. 3xTg‐AD −2.42, p = 0.046] females [t(8,9) and −2.42, p = VWR vs.= 3xTg-AD 0.046] males and vs. [t(7,8) = −3.14, 3xTg‐AD VWR p =males 0.021].[t(7,8) = −3.14, In the muscularp = 0.021]. enduranceIn thetrial a muscular days effectendurance [D, F(1,30)trial a days = 13.98, p