- Email: [email protected]
Progressive age-dependent motor impairment in human tau P301S overexpressing mice
Behavioural Brain Research 376 (2019) 112158
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Progressive age-dependent motor impairment in human tau P301S overexpressing mice
T
Hennariikka Koivistoa, Ellen Ytebrouckb, Sofie Carmansb, Reyhaneh Naderic, Pasi O. Miettinena, ⁎ Bart Roucourtb, Heikki Tanilaa, a
A. I. Virtanen Institute, University of Eastern Finland, Kuopio, Finland reMYND, Leuven, Belgium c Department of Biology, Shahid Bahonar University, Kerman, Iran b
A R T I C LE I N FO
A B S T R A C T
Keywords: Tauopathy Alzheimer’s disease Frontotemporal dementia Motor coordination Gait Cerebellar nuclei
This study assessed the development of motor deficits in female hTau.P301S transgenic mice from 1.5 to 5.5 months of age. The test battery included clasping reflex, grid hanging, Rotarod test, spontaneous explorative activity, Catwalk gait analysis, and nest building. Starting from the age of 2–3 months the mice showed marked hyperactivity, abnormal placing of weight on the hindlimbs and defective nest building in their home cage. These behavioral impairments did not progress with age. In addition, there was a progressive development of hindlimb clasping, inability to stay on a rotating rod or hang on a metal grid, and gait impairment. Depending on the measured output parameter, the motor impairment became significant from 3 to 4 months onwards and rapidly worsened until the age of 5.5 months with little inter-individual variation. The progressive motor impairment was paralleled by a robust increase in AT8 p-tau positive neurons in deep cerebellar nuclei and pontine brainstem between 3 and 5.5 months of age. The quick and steadily progressive motor impairment between 3 and 5.5 months of age accompanied by robust development of tau pathology in the hindbrain makes this mouse well suited for preclinical studies aiming at slowing down tau pathology associated with primary or secondary tauopathies.
1. Introduction Abnormal phosphorylation and aggregation of microtubule-associated protein tau are the characteristic neuropathological features of several neurodegenerative diseases of an old age, including frontotemporal lobar degeneration, Pick’s disease, supranuclear palsy and corticobasal degeneration. In addition to these so-called tauopathies, neurofibrillary tangles comprising of paired helical tau filaments are an essential component of Alzheimer neuropathology. Frontotemporal dementia with parkinsonism related to chromosome 17 (FTDP-17) is perhaps the best-known example of a familiar tauopathy and the basis of various transgenic mouse models used in the research of tau pathology. The P301S tau mutation is associated with an aggressive earlyonset and rapidly progressive form of FTDP-17 [1]. A transgenic mouse expressing human tau with the P301S mutation under the control of the murine thy1 promoter, referred to as the hTau.P301S mouse, was first described in 2002 [2]. At 5–6 months of age, homozygous animals from this line developed a neurological phenotype that was characterized by general muscle weakness, tremor and a severe paraparesis. At this age,
⁎
numerous phospho-tau positive neurons were see throughout the brain, most abundantly in the spinal cord and brainstem [2]. The hTau.P301S mouse when backcrossed to C57Bl/6 J x CBA/ca background, presents rapidly progressing tau pathology with little variation between individuals, which makes is well suited for assessment of anti-tau treatments. Despite its use in several preclinical trials over the years, including in a large tau immunization trial [3], there is no published systematic study on the development of motor symptoms in this mouse model. One study has focused on the behavioral markers in young (1–4 month) hTau.P301S mice and found impaired performance in the accelerated Rotarod starting after 3 months of age [4], but did not follow up any other motor output parameter. Therefore, we set out to determine the time course of motor impairment in hTau.P301S mice between 1.5 and 5.5 months of age (i.e.before the age of general deterioration) to establish a reference for future preclinical treatment trials employing this mouse model.
Corresponding author at: A. I. Virtanen Institute, University of Eastern Finland, P.O. Box 1627, 70211, Kuopio, Finland. E-mail address: Heikki.Tanila@uef.fi (H. Tanila).
https://doi.org/10.1016/j.bbr.2019.112158 Received 4 April 2019; Received in revised form 13 August 2019; Accepted 17 August 2019 Available online 20 August 2019 0166-4328/ © 2019 Elsevier B.V. All rights reserved.
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
carefully placed upside down as the lid of a 24 cm x 35.5 cm x 24 cm transparent plastic cage. The time to fall off until a cut-off time of 300 s was measured with a stop-watch. The test was repeated three times with a 10-min interval between the trials, and the best result was recorded.
Table 1 Four mouse cohorts used in the study. F = female, TG = hTau.P301S, WT = wild-type control. Age
Sex
TG(N)
WT(N)
Tests
2.5 – 6 mo 2.5 – 5.5 mo 1.5 – 3.5 mo
F F F
10 18 15
10 19 10
2.5 – 5.5 mo
F
10
10
Body weight, clasping score weekly Rotarod every two weeks Grid hanging, nest building once per month Explorative activity, Catwalk once per month
2.2.4. Catwalk gait analysis Possible gait alternations were measured by CatWalk XT (Noldus, Wageningen, the Netherlands), an automated gait analysis system. Mice were let run freely through a 130 cm long alley with a glass plate floor. A high-speed camera recorded the paw contacts from below, and the program analyzed numerous static and dynamic parameters assessing individual paw functioning and gait patterns. We recorded 3–5 uninterrupted runs through the alley and selected the following parameters for analysis: print area, stride length, and swing speed for each paw, and base of support for forepaws and hindpaws. These parameters were initially calculated for each run and for each paw, then averaged over the runs, and finally the values for the left and right front and hind paws were averaged.
2. Methods 2.1. Animals Generation of the hTau.P301S transgenic mice was originally described by Allen et al. [2]. Mice were kept in their original C57BL/6 J x CBA/ca background and bred as a homozygous strain. C57BL/6 J x CBA/ca F1 hybrid mice were used as wild-type controls. Because of aggressive behavior, male hTau.P301S mice are difficult to maintain in large quantities and have seldom been used in behavioral studies. Therefore, we focused on female mice only. Four separate cohorts of mice were used in the study to avoid confounding cross-test and multiple testing effects (see Table 1 for details). The animals were kept group-housed in standard laboratory rodent cages under control conditions with a 12 h/12 h dark-light cycle. All experiments were run during the light phase. Food and water were provided ad libitum. The mice were weighed and their general health status was recorded once a week before 3 months of age and then three times per week. All experiments were conducted in accordance with the EU directive 2010/ 63/EU for animal well-being, using protocols approved by the Animal Experiment Board of Finland and by Ethical committee (operating under approval number LA1210532) in Belgium.
2.2.5. Spontaneous explorative activity Spontaneous explorative activity was assessed by using an automated activity monitor (TruScan, Coulbourn Instruments, Whitehall, PA, USA) based on infrared photobeam detection. The system consisted of an observation cage with white plastic walls (26 cm x 26 cm x 39 cm) and two frames of photo detectors enabling separate monitoring of horizontal (XY-move time) and vertical activity (rearing). The test cage was cleaned with 70% ethanol before each mouse to avoid odor traces. The following parameters were measured during a 10-min session: ambulatory distance (gross horizontal locomotion) and rearing time. 2.2.6. Nest building This test was used to assess execution of goal directed motor sequences and was done according to the original description by its inventor Robert Deacon [5] with small modifications. The test was conducted in the home cage. A 22 cm × 22 cm sheet of tissue paper was left in the cage overnight. Next morning the shredding of the material and building of a nest-like construct was evaluated on a 5-point scale as follows: 0 = nothing done (> 90% of the material untouched) 1 = paper partially shredded (50–90% remaining intact) 2 = paper mostly torn up (50–90%) but no identifiable nest is built, pieces are outspread 3 = an identifiable but flat nest: > 90% of the paper is torn up, nest walls are higher than mouse height but in less than 50% of the nest circumference 4 = complete nest, > 90% of paper is torn up, the nest walls are higher than mouse height on more than 50% of the nest circumference
2.2. Behavioral protocols 2.2.1. Clasping To score the clasping reflex, mice were kept approximately 1.5 cm above their tail base for about 10 s. Clasping of the hind limbs was scored for each limb separately (0 = hind limb folded backwards and distant from body, angle of limb with body axis > 0°; 0.5 = any phenotype in between 0 and 1; 1 hind limb partially retracted during more than 50% of the time; 3 = limb completely retracted during more than 50% of the time). 2.2.2. Rotarod The Rotarod test was performed to assess motor coordination and balance deficits. Mice were tested in an automated Rotarod device (Model: LE-8200, Five Station Accelerating Rotarod, Bioseb, France), with a maximum of five mice at a time placed on a revolving rod with opaque side walls. The first Rotarod sessions at 2.5 months of age was performed on four consecutive days (familiarization on day 1, four trainings sessions at a constant speed of 10 rpm for a maximum time of 120 s on day 2, four consecutive test trials at 10 rpm for 120 s on day 3, and two acceleration tests wherein the speed of the rod increased from 4 to 40 rpm within 120 s on day 4). At 3 and 4 months of age, the familiarization was omitted. From 4.5 months onwards the training sessions were omitted as well. During the test trials, the mean latency to fall off the Rotarod was recorded and used in the analysis. For the acceleration test, the mean maximum speed and the mean latency to fall off the rod were recorded and analyzed.
2.3. Histopathology Mice were deeply anesthetized with a pentobarbital/chroralhydrate cocktail (60 mg/kg each) and perfused transcardially with ice-cold 0.9% saline followed by 4% paraformaldehyde. After 4 h of postfixation, brains were incubated overnight at +4 °C in 30% sucrose solution and finally stored in a cryoprotectant at −20 °C before sectioning on a freezing microtome into 35 μm thick coronal sections. To visualize pathological tau aggregates, sections were stained for monoclonal anti-human PHF-Tau, clone AT8 (1:1000, Thermo Scientific, IL, USA). After citrate solution pretreatment, sections were treated with 2% hydrogen peroxidase in methanol. After rinsing sections were blocked in 3% bovine serum albumin in TBS-T for 60 min. The primary antibody was also diluted into 3% BSA in TBS-T. The sections were incubated overnight at 4 °C on a shaker table. Following incubation, the sections were rinsed thoroughly with TBS-T and transferred to a solution containing the secondary antibody, goat–anti mouse labelled with Alexa Fluor 488, 1:400 (Invitrogen, Life Technologies
2.2.3. Grid hanging This test was used to determine static muscle force. The mouse was placed on a 20 cm x 25 cm wire grid (grid unit 1 cm by 1 cm) that was 2
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Corporation, OR, USA). After 2-h incubation the sections were rinsed three times and embedded on slides and mounted with Vectashield (Vector Hard Set –mounting medium with DAPI, Vector Laboratories, CA, USA). The sections were photographed using Axio Imager M2 microscope (Zeiss, Jena, Germany) with fluorescent AxioCam MR3 camera attached. The pictures were then examined with Zen pro 2011 blue edition (Zeiss, Jena, Germany).
2.4. Statistics The statistical analyses were done using either Graph Pad Prism 5 or IBM SPSS Statistics 21 software. Analysis of variance for repeated measures (ANOVA-RM) was done for the initial genotype comparison, with the age of the mice as the within-subject and the genotype as the between-subject factor. However, for clasping we used one-sample ttest, since none of the wild-type mice showed the clasping reflex. In addition, since some parameters of the Catwalk gait analysis, such as the paw print areas are dependent on the body weight, we also ran the ANOVA-RM with the body weight as a covariant for these parameters. Depending on the parameter distributions, the post-hoc tests were done with the Least Significant Difference (LSD) test, one-sample t-test or nonparametric Mann-Whitney test. The level of statistical significance was set at p < 0.05.
Fig. 2. Clasping score development between hTau.P301S (hTau) and wild-type (WT) mice during ages 3–6 months. *p < 0.05, **p < 0.01, ***p < 0.001, one-sample t-test. The error bars denote SEMs.
3.3. Impaired muscle coordination and static force in hTau.P301S mice at 4 months of age To assess motor coordination and balance, we used the common Rotarod test. Due to hyperactivity at a young age, hTau.P301S mice needed an extended period of adaptation to the Rotarod test (which was also applied to WT controls) but adapted to repeated testing without further training. We tested the mice in two versions of the test, with a fixed and accelerating speed of rotation. There was an overall genotype × age interaction in both modes (fixed speed: F5,22 = 17.1, p < 0.001; accelerating speed: F5,22 = 27.8, p < 0.001), such that hTau.P301S mice showed a steady decline from 3 months on, reaching significance at 4 months (Fig. 3AB). To assess plain static force with little demand on muscle coordination, we used the grid hanging test for a separate group of young mice. In this test, none of the mice fell before the 5 min cut-off time at 2.5 months of age, but a significant decline appeared in hTau.P301S mice at 3.5 months (Fig. 3C).
3. Results 3.1. Body weight declines in hTau.P301S mice after 4 months of age There was an overall genotype difference in the body weight (F1,18 = 28.5, p < 0.001) and an age × genotype interaction (F38,684 = 82.6, p < 0.001). However, a closer look at the development of body weight suggested that 4 months of age is a turn point for the hTau.P301S mice (Fig. 1). Before the age of 4 months, the body weights of hTau.P301S mice did not differ significantly from WT mice (F1,18 = 3.2, p = 0.09), but showed a decline thereafter while a steady growth continued in WT mice.
3.4. Gait analysis reveals altered posture in hTau.P301S mice as early as at 2.5 months of age To assess the earliest signs of motor impairment, we used the automated Catwalk gait analysis in a separate group of young mice from 2.5 to 4.5 months of age. Fig. 4 shows an example of the automatically assigned footprints. Across these ages, hTau.P301S mice showed an altered posture. The print area of forepaws was significantly smaller in hTau.P301S mice than in WT mice (F1,17 = 22.8, p < 0.001; Fig. 5A), while there was no genotype difference in hindpaws (F1,17 = 1.4, p = 0.25; Fig. 5B). The genotype difference in the forepaw print area remained significant even after correction for the body weight (F1,16 = 7.4, p = 0.015). Further, the base of support (distance between the paws) for forepaws was wider (Fig. 5C) and that of hindpaws narrower in hTau.P301S mice compared to WT mice (Fig. 5D). The difference could be seen already at 2.5 months of age, but clearly accentuated after 3.5 months, which resulted in a significant genotype × age interaction (forepaws: F2,32 = 15.5, p < 0.001; hindpaws: F2,32 = 13.4, p < 0.001). The genotype difference in the base of support of the forepaws disappeared after weight correction (F1,15 = 2.3, p = 0.15) but the genotype × age interaction remained significant (F2,30 = 9.0, p = 0.001). In contrast, both the genotype difference (F1,15 = 10.5, p = 0.006) and the genotype × age interaction (F2,14 = 6.0, p = 0.013) remained significant for the hindpaws after the weight correction. These observations indicate that hTau.P301S mice placed their weight more on hindlimbs. Difficulties in maintaining the balance was also reflected in the swing time of the limbs during the step cycle. The swing time of forelimbs as % of the step cycle showed gradual shortening in hTau.P301S mice as they aged (Fig. 5E), resulting in
3.2. Clasping of the hindpaws appears in hTau.P301S mice after 4 months of age Clasping of the hindpaw was not present in any of the WT mice. However, the clasping score showed a steady increase in hTau.P301S mice from 4 months of age onwards until reaching an asymptote after 5 months (Fig. 2).
Fig. 1. Body weight gain from 2.5 to 6 months of age in hTau.P301S (hTau) and wild-type (WT) mice. Note the clear deviation of the growth curves of the genotypes around 4 months of age. *p < 0.05, **p < 0.01, ***p < 0.001, post-hoc LSD-test for triplet age points. The error bars denote SEMs. 3
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Fig. 3. Tests for balance/dynamic muscle force and static muscle force. (A) Rotarod test with fixed speed (10 rpm). (B) Rotarod test with acceleration. (C) Grid hanging test. The mean ± SEM latency to fall is shown. *p < 0.05, **p < 0.01, ***p < 0.001; A and B, post-hoc LSD-test, C Mann-Whitney test.
5.5 months of age (Fig. 6A). This trend led to a significant genotype × age interaction in the ambulatory distance moved in a fixed time in the test environment (F3,42 = 13.1, p < 0.001). However, the time spend in rearing, which is another indicator of exploratory activity but requires much more balancing, was lower in hTau.P301S than WT mice at all studied ages (genotype main effect: F1,14 = 10.8, p = 0.005; interaction. F3,42 = 0.9, p = 0.47; Fig. 6B).
3.6. Robust impairment in nest building of hTau.P301S mice at all studied ages Finally, we wanted to assess the ability of hTau.P301S mice to perform a complex motor sequence. For such we chose nest building, which is part of the natural behavioral repertoire of mice and does not need additional training. It was obvious that the nest scores remained substantially lower in hTau.P301S mice than their WT controls across all ages from 2.5 to 5.5 months (F1,14 = 138.3, p < 0.001; Fig. 6C). Fig. 4. Example of footprints on the Catwalk apparatus. (Top) A 2.5-month-old wild-type mouse, (bottom) a hTau.P301S transgenic mouse of the same age. LH = left hindpaw, LF = left forepaw, RH = right hindpaw, RF = right forepaw. The hindpaw prints are marked with a red rectangle.
3.7. Earliest signs of pTau aggregation in the brain stem are seen in pontine nuclei To determine a structure-function relationship between the tau pathology and motor impairment as a function of age, we examined the location of pathological hTau aggregation in the brain with the common AT8 antibody recognizing the phosphorylation of serine 202 and threonine 205 of hTau. The spinal cord pathology in the hTau.P301S mouse model has been described earlier [2]. The basal ganglia or cerebellar hemispheres showed no significant AT8 positivity at any age. In contrast, we found substantial aggregation of AT8 positive neurons in the deep cerebellar and medial vestibular nuclei in the pons as early as at 3 months of age (Fig. 7A). There was also scattered AT8 positivity in the lateral vestibular, pontine reticular and gigantocellular nuclei. At 5.5 months, the entire brainstem was full of AT8 positive aggregates (Fig. 7B).
a significant genotype × age interaction (F2,16 = 14.9, p < 0.001), while similar shortening of the hindlimb swing time was present in hTau.P301S mice across ages (Fig. 5F) (genotype main effect: F2,16 = 18.5, p < 0.001; genotype × age interaction: F2,16 = 1.3, p = 0.30). 3.5. Bidirectional change in spontaneous locomotor activity of hTau.P301S mice across ages Young hTau.P301S mice were hyperactive compared to their WT controls. However, as their motor abilities started to decline, their activity level first ostensibly ‘normalized’, followed by hypolocomotion at 4
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Fig. 5. Catwalk gait analysis of hTau.P301S (hTau) and wild-type (WT) mice. (A) Footprint area of forepaws and (B) hindpaws. (C) Base of support (the distance between the supporting pair of paws) for forepaws and (D) hindpaws. (E) Duration of swing (%) of the step cycle for forepaws and (F) hindpaws. *p < 0.05, **p < 0.01, ***p < 0.001; A and B, post-hoc LSD-test. The error bars denote SEMs. Fig. 6. Spontaneous exploratory activity of hTau.P301S (hTau) and wild-type (WT) mice in the TruScan measured with infrared photobeam detectors. (A) Ambulatory distance (gross horizontal locomotion), (B) rearing time. (C) Nest building score on the scale 0 – 4. *p < 0.05, **p < 0.01, ***p < 0.001; A and B, post-hoc LSD-test, C, Mann-Whitney test. The error bars denote SEMs.
5
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Fig. 7. (A) P-tau aggregations in female hTau.P301S mice at 3 months of age. LEC = lateral entorhinal cortex; PrC = perirhinal cortex; DN = dentate, IpN = interposed, FgN = fastigial cerebelar nucleus; mVe, spVe = medial, spinal vestibular nucleus; PnC = pontine reticular nucleus, caudal part; GiN = gigantocellular nucleus. (B). P-tau aggregations in female hTau.P301S mice at 5.5 months of age. mEC = medial entorhinal cortex,PSub = parasubiculum; dlPAG, laPAG = dorsolateral, lateral periaqueductal gray matter; InCo = intercollicular nucleus; CnF = cuneiform n. ; PnO = pontine reticular n., oral part; PnC = pontine reticular n., caudal part; m5 = motor trigeminal n. ; sp5 = spinal trigeminal n.; 7 = facial nucleus. The coordinates above the sections are from bregma. Scale bar =1.0 mm.
4. Discussion
The early hyperactivity appears to be specific for the hTau.P301S mouse line. Consistent with present findings, a previous study on the early behavioral phenotype of this mouse line also documented hyperlocomotion at 2 months of age [4]. Hyperlocomotion was also observed in the other common mouse line expressing P301S hTau under the mouse prion promoter (aka PS19 line, [6]), but it showed progression with age and became significant only at the age of 7 months [7]. Further, a recently described third P301S hTau mouse (line Tau 58/2; [8]) showed no hyperlocomotion before 12 months of age, albeit axonal pathology and p-tau positive neurons could be observed throughout the brain as early as 3 months of age and impaired Rotarod performance at 6 months [9]. Thus, it appears that the hyperactivity is not directly related to accumulation of hyperphosphorylated tau in the brain of these mouse lines, although hyperlocomotion related to overexpression of mutated human amyloid precursor protein in mice could be reversed by deletion of mouse endogenous tau in one study [10]. The age of 4 months appears to be a watershed for the phenotype of the hTau.P301S mouse. There was a significant decline in body weight
To our knowledge, this is the first systematic study on the development of motor deficits in the hTau.P301S transgenic mouse across the adult life span. We observed two patterns of deficits with different progression with age. Already at the earliest ages studied, 2–3 months, the mice showed marked hyperactivity, abnormal placing of weight on the hindlimbs and defective nest building in their home cage. These behavioral impairments showed no progression with age. The other pattern consisted of progressive development of hindlimb clasping, inability to stay on a rotating rod or hang on a metal grid, and gait impairment. Depending on the measured output parameter, the progressive motor impairment became significant from 3 to 4 months onwards and showed a steep linear trend until the age of 5.5 months with little inter-individual variation. The progressive motor impairment was paralleled by a robust increase in AT8 p-tau positive neurons in deep cerebellar nuclei and pontine brainstem between 3 and 5.5 months of age. 6
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
or hindpaws. Of note, hindlimb paralysis appeared in this mouse at 6 months of age as in the hTau.P301S mouse. We observed slowing of hindlimb swing speed as well, but not before 5.5 months (data now shown). However, the parameters sensitive to body posture (print area and base of support of fore- and hindpaws) and prolongation of the stand phase of the step cycle were abnormal already at earlier ages. We recently published a Catwalk analysis of a mouse with selective degeneration of cerebellar Purkinje cells and secondary lesions in the deep cerebellar nuclei [16]. This mouse showed progressive shortening of stride length and slowing of swing time of both limbs, as well as widening of the forepaw base of support, thus showing partially overlapping pattern of both hTau mouse models. This finding thus lends support to the notion that dysfunction of deep cerebellar nuclei contribute to the observed gait deficit. To conclude, the hTau.P301S mouse demonstrates a fast and steadily progressive motor impairment between 3 and 5.5 months of age. The impairment can be easily monitored by the hindpaw clasping response and a small battery of standard motor tests. During this time window, the motor impairment is paralleled by robust development of tau pathology in deep cerebellar nuclei and pontine brainstem, which is easy to localize in coronal brain sections around the 4th ventricle. This mouse is thus well suited for preclinical studies aiming at slowing down tau pathology before the symptomatic phase.
at 4 months, which is also the age when many indicators of motor impairment show a significant change. The time course of decline in the body weight and motor functions raise the question whether the motor impairment may result from, for instance, decreased muscle mass. We did not measure the muscle mass directly but found no indication of atrophy of special muscle groups. The most direct measure of static muscle force in our test battery was the grid hanging task where impairment of hTau P301S mice was evident as early as 3.5 months of age, i.e.two weeks before the body weight started to decline (Fig. 3C). Another parameter sensitive to muscle force was explorative rearing. Indeed, the rearing time showed a decline in hTauP301S mice between 3.5 and 4.5 months of age, but for some reason a similar decline was also seen in WT mice (Fig. 6B). On the other hand, the clasping reflex appeared in hTau P301S mice precisely after 4 months of age but it is totally independent of muscle force. Taken together, muscle weakness per se cannot explain the observed motor phenotype. The abnormal posture of hTau.P301S mice, with increased weight on hindlimbs and wider base of support of the forelimbs, may be explained by early tau pathology in lumbal spinal motoneurons. However, the abnormal posture was present already at 2.5 months of age when there was no evidence of hindlimb weakness in terms of static muscle force as assessed in the grid hanging test. Furthermore, the relative shortening of the swing during the step cycle in hTau.P301S mice in fact means that the swing speed was increased. This fits poorly with dysfunction at the spinal level. Another, and more plausible, explanation for the abnormal posture is early tau pathology in the brainstem. Namely, both lateral vestibular nucleus and pontine reticular formation send projections to the lumbal spinal motoneurons and thereby influence the muscle tone of the hindlimbs and body posture [11]. Recent evidence suggests that stimulation of the ventral part of the mesencephalic locomotor center in the pontine reticular formation induces rearing in rats, while it lesioning results in atonia [12]. Impairment of this brainstem neuron group that maintains the tonus of the hindlimb extensors may well account for the abnormal posture and decreased rearing of hTau.P301S mice despite hyperactivity, which usually involves rearing besides horizontal locomotion. The most striking sign of abnormality in hTau.P301S mice was clasping of hindpaws when the mouse was lifted by the tail. It started consistently in all mice around 4 months of age and reached it full extent by 5 months of age. Similar hindpaw clasping was reported in the PS19 mouse starting from 3 months of age [6]. The clasping response has been reported in several genetically modified mouse lines, but most consistently in mice with primary pathology in the cerebellum or striatum [13]. Since the hTau.P301S mouse did not express tau pathology in the striatum, and deep cerebellar nuclei constitute the only output channel of cerebellum, the observed pathological tau aggregations in fastigial and interposite cerebellar nuclei are likely candidates to account for the abnormal hindpaw clasping. Further, the rapid progression of intracellular p-tau aggregation in this brain region matches the appearance of the clasping response. In agreement with the report on early behavioral abnormalities hTau.P301S mice [4] we found significant impairment in the Rotarod performance at 4 months of age. Since successful performance in this task requires both motor coordination and balance, it is sensitive to a variety of lesions. Typically, genetic mouse models of cerebellar ataxia display impairment in the Rotarod test [14]. In addition, the vestibular nuclei play a key role in maintaining the balance. Thus, the observation of early p-tau deposits in the fastigial cerebellar nucleus and vestibular nuclei at 3 months, and their rapid progression thereafter, likely contribute to this deficit. The Catwalk gait analysis has been employed in only one study on a tauopathy mouse so far [15]. A mouse expressing truncated human tau and tau aggregation mainly in the brainstem showed no significant deviance from the controls at 3 and 4 months of age, but at 5 months the stride length became shorter and swing time slower; however, the body posture was not changed, as evidenced by print areas of the fore-
Declaration of Competing Interest BR and EY and SC are employees of reMYND N.V. Acknowledgement The study was supported by EU-Horizon 2020 project PANA (grant 686009-2). References [1] A.D. Sperfeld, M.B. Collatz, H. Baier, M. Palmbach, A. Storch, J. Schwarz, K. Tatsch, S. Reske, M. Joosse, P. Heutink, A.C. Ludolph, FTDP-17: an early-onset phenotype with parkinsonism and epileptic seizures caused by a novel mutation, Ann. Neurol. 46 (Nov (5)) (1999) 708–715. [2] B. Allen, E. Ingram, M. Takao, M.J. Smith, R. Jakes, K. Virdee, H. Yoshida, M. Holzer, M. Craxton, P.C. Emson, C. Atzori, A. Migheli, R.A. Crowther, B. Ghetti, M.G. Spillantini, M. Goedert, Abundant tau filaments and nonapoptotic neurodegeneration in transgenic mice expressing human P301S tau protein, J. Neurosci. 22 (2002) 9340–9351. [3] X. Chai, S. Wu, T.K. Murray, R. Kinley, C.V. Cella, H. Sims, N. Buckner, J. Hanmer, P. Davies, M.J. O’Neill, M.L. Hutton, M. Citron, Passive immunization with anti-Tau antibodies in two transgenic models: reduction of Tau pathology and delay of disease progression, J. Biol. Chem. 286 (Sep (39)) (2011) 34457–34467. [4] M.L. Scattoni, L. Gasparini, E. Alleva, M. Goedert, G. Calamandrei, M.G. Spillantini, Early behavioural markers of disease in P301S tau transgenic mice, Behav. Brain Res. 208 (Mar (1)) (2010) 250–257. [5] R.M. Deacon, Assessing nest building in mice, Nat. Protoc. 1 (3) (2006) 1117–1119. [6] Y. Yoshiyama, M. Higuchi, B. Zhang, S.M. Huang, N. Iwata, T.C. Saido, J. Maeda, T. Suhara, J.Q. Trojanowski, V.M. Lee, Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model, Neuron 53 (Feb (3)) (2007) 337–351. [7] M. Gerges, M. Dumont, C. Stack, C. Elipenahli, S. Jainuddin, N.N. Starkova, L. Yang, A.A. Starkov, F. Beal, Behavioral deficit, oxidative stress, and mitochondrial dysfunction precede tau pathology in P301S transgenic mice, FASEB J. 25 (2011) 4063–4072. [8] J. van Eersel, C.H. Stevens, M. Przybyla, A. Gladbach, K. Stefanoska, C.K. Chan, W.Y. Ong, J.R. Hodges, G.T. Sutherland, J.J. Kril, D. Abramowski, M. Staufenbiel, G.M. Halliday, L.M. Ittner, Early-onset axonal pathology in a novel P301S-Tau transgenic mouse model of frontotemporal lobar degeneration, Neuropathol. Appl. Neurobiol. 41 (2015) 906–925. [9] M. Kassiou, M. Przybyla, C.H. Stevens, J. van der Hoven, A. Harasta, M. Bi, A. Ittner, A. van Hummel, J.R. Hodges, O. Piguet, T. Karl, G.D. Housley, Y.D. Ke, L.M. Ittner, J. Eersel, Disinhibition-like behavior in a P301S mutant tau transgenic mouse model of frontotemporal dementia, Neurosci. Lett. 631 (2016) 24–29. [10] M. Michikawa, M. Yoshikawa, Y. Soeda, O.F.X. Almeida, A. Takashima, Tau depletion in APP transgenic mice attenuates task-related hyperactivation of the Hippocampus and differentially influences locomotor activity and spatial memory, Front. Neurosci. 12 (2018) 124. [11] T. Terashima, K. Inoue, Y. Inoue, K. Mikoshiba, Y. Tsukada, Observations on the brainstem-spinal descending systems of normal and reeler mutant mice by the
7
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
[15] I. Zimova, V. Brezovakova, T. Hromadka, P. Weisova, V. Cubinkova, B. Valachova, P. Filipcik, S. Jadhav, T. Smolek, M. Novak, N. Zilka, Human truncated tau induces mature neurofibrillary pathology in a mouse model of human tauopathy, J. Alzheimers Dis. 54 (2016) 831–843. [16] R.R. Nair, H. Koivisto, K. Jokivarsi, I.J. Miinalainen, K.J. Autio, A. Manninen, P. Poutiainen, H. Tanila, J.K. Hiltunen, A.J. Kastaniotis, Impaired mitochondrial fatty acid synthesis leads to neurodegeneration in mice, J. Neurosci. 38 (2018) 9781–9800.
retrograde HRP method, J. Comp. Neurol. 225 (1984) 95–104. [12] D. Sherman, P.M. Fuller, J. Marcus, J. Yu, P. Zhang, N.L. Chamberlin, C.B. Saper, J. Lu, Anatomical location of the mesencephalic locomotor region and its possible role in locomotion, posture, cataplexy, and parkinsonism, Front. Neurol. 6 (2015) 140. [13] R. Lalonde, C. Strazielle, Brain regions and genes affecting limb-clasping responses, Brain Res. Rev. 67 (2011) 252–259. [14] J.N. Crawley, What’s Wrong with My Mouse? 2nd ed., John Wiley & Sons, Hoboken, NJ, USA, 2007, pp. 80–81.
8
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Progressive age-dependent motor impairment in human tau P301S overexpressing mice
T
Hennariikka Koivistoa, Ellen Ytebrouckb, Sofie Carmansb, Reyhaneh Naderic, Pasi O. Miettinena, ⁎ Bart Roucourtb, Heikki Tanilaa, a
A. I. Virtanen Institute, University of Eastern Finland, Kuopio, Finland reMYND, Leuven, Belgium c Department of Biology, Shahid Bahonar University, Kerman, Iran b
A R T I C LE I N FO
A B S T R A C T
Keywords: Tauopathy Alzheimer’s disease Frontotemporal dementia Motor coordination Gait Cerebellar nuclei
This study assessed the development of motor deficits in female hTau.P301S transgenic mice from 1.5 to 5.5 months of age. The test battery included clasping reflex, grid hanging, Rotarod test, spontaneous explorative activity, Catwalk gait analysis, and nest building. Starting from the age of 2–3 months the mice showed marked hyperactivity, abnormal placing of weight on the hindlimbs and defective nest building in their home cage. These behavioral impairments did not progress with age. In addition, there was a progressive development of hindlimb clasping, inability to stay on a rotating rod or hang on a metal grid, and gait impairment. Depending on the measured output parameter, the motor impairment became significant from 3 to 4 months onwards and rapidly worsened until the age of 5.5 months with little inter-individual variation. The progressive motor impairment was paralleled by a robust increase in AT8 p-tau positive neurons in deep cerebellar nuclei and pontine brainstem between 3 and 5.5 months of age. The quick and steadily progressive motor impairment between 3 and 5.5 months of age accompanied by robust development of tau pathology in the hindbrain makes this mouse well suited for preclinical studies aiming at slowing down tau pathology associated with primary or secondary tauopathies.
1. Introduction Abnormal phosphorylation and aggregation of microtubule-associated protein tau are the characteristic neuropathological features of several neurodegenerative diseases of an old age, including frontotemporal lobar degeneration, Pick’s disease, supranuclear palsy and corticobasal degeneration. In addition to these so-called tauopathies, neurofibrillary tangles comprising of paired helical tau filaments are an essential component of Alzheimer neuropathology. Frontotemporal dementia with parkinsonism related to chromosome 17 (FTDP-17) is perhaps the best-known example of a familiar tauopathy and the basis of various transgenic mouse models used in the research of tau pathology. The P301S tau mutation is associated with an aggressive earlyonset and rapidly progressive form of FTDP-17 [1]. A transgenic mouse expressing human tau with the P301S mutation under the control of the murine thy1 promoter, referred to as the hTau.P301S mouse, was first described in 2002 [2]. At 5–6 months of age, homozygous animals from this line developed a neurological phenotype that was characterized by general muscle weakness, tremor and a severe paraparesis. At this age,
⁎
numerous phospho-tau positive neurons were see throughout the brain, most abundantly in the spinal cord and brainstem [2]. The hTau.P301S mouse when backcrossed to C57Bl/6 J x CBA/ca background, presents rapidly progressing tau pathology with little variation between individuals, which makes is well suited for assessment of anti-tau treatments. Despite its use in several preclinical trials over the years, including in a large tau immunization trial [3], there is no published systematic study on the development of motor symptoms in this mouse model. One study has focused on the behavioral markers in young (1–4 month) hTau.P301S mice and found impaired performance in the accelerated Rotarod starting after 3 months of age [4], but did not follow up any other motor output parameter. Therefore, we set out to determine the time course of motor impairment in hTau.P301S mice between 1.5 and 5.5 months of age (i.e.before the age of general deterioration) to establish a reference for future preclinical treatment trials employing this mouse model.
Corresponding author at: A. I. Virtanen Institute, University of Eastern Finland, P.O. Box 1627, 70211, Kuopio, Finland. E-mail address: Heikki.Tanila@uef.fi (H. Tanila).
https://doi.org/10.1016/j.bbr.2019.112158 Received 4 April 2019; Received in revised form 13 August 2019; Accepted 17 August 2019 Available online 20 August 2019 0166-4328/ © 2019 Elsevier B.V. All rights reserved.
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
carefully placed upside down as the lid of a 24 cm x 35.5 cm x 24 cm transparent plastic cage. The time to fall off until a cut-off time of 300 s was measured with a stop-watch. The test was repeated three times with a 10-min interval between the trials, and the best result was recorded.
Table 1 Four mouse cohorts used in the study. F = female, TG = hTau.P301S, WT = wild-type control. Age
Sex
TG(N)
WT(N)
Tests
2.5 – 6 mo 2.5 – 5.5 mo 1.5 – 3.5 mo
F F F
10 18 15
10 19 10
2.5 – 5.5 mo
F
10
10
Body weight, clasping score weekly Rotarod every two weeks Grid hanging, nest building once per month Explorative activity, Catwalk once per month
2.2.4. Catwalk gait analysis Possible gait alternations were measured by CatWalk XT (Noldus, Wageningen, the Netherlands), an automated gait analysis system. Mice were let run freely through a 130 cm long alley with a glass plate floor. A high-speed camera recorded the paw contacts from below, and the program analyzed numerous static and dynamic parameters assessing individual paw functioning and gait patterns. We recorded 3–5 uninterrupted runs through the alley and selected the following parameters for analysis: print area, stride length, and swing speed for each paw, and base of support for forepaws and hindpaws. These parameters were initially calculated for each run and for each paw, then averaged over the runs, and finally the values for the left and right front and hind paws were averaged.
2. Methods 2.1. Animals Generation of the hTau.P301S transgenic mice was originally described by Allen et al. [2]. Mice were kept in their original C57BL/6 J x CBA/ca background and bred as a homozygous strain. C57BL/6 J x CBA/ca F1 hybrid mice were used as wild-type controls. Because of aggressive behavior, male hTau.P301S mice are difficult to maintain in large quantities and have seldom been used in behavioral studies. Therefore, we focused on female mice only. Four separate cohorts of mice were used in the study to avoid confounding cross-test and multiple testing effects (see Table 1 for details). The animals were kept group-housed in standard laboratory rodent cages under control conditions with a 12 h/12 h dark-light cycle. All experiments were run during the light phase. Food and water were provided ad libitum. The mice were weighed and their general health status was recorded once a week before 3 months of age and then three times per week. All experiments were conducted in accordance with the EU directive 2010/ 63/EU for animal well-being, using protocols approved by the Animal Experiment Board of Finland and by Ethical committee (operating under approval number LA1210532) in Belgium.
2.2.5. Spontaneous explorative activity Spontaneous explorative activity was assessed by using an automated activity monitor (TruScan, Coulbourn Instruments, Whitehall, PA, USA) based on infrared photobeam detection. The system consisted of an observation cage with white plastic walls (26 cm x 26 cm x 39 cm) and two frames of photo detectors enabling separate monitoring of horizontal (XY-move time) and vertical activity (rearing). The test cage was cleaned with 70% ethanol before each mouse to avoid odor traces. The following parameters were measured during a 10-min session: ambulatory distance (gross horizontal locomotion) and rearing time. 2.2.6. Nest building This test was used to assess execution of goal directed motor sequences and was done according to the original description by its inventor Robert Deacon [5] with small modifications. The test was conducted in the home cage. A 22 cm × 22 cm sheet of tissue paper was left in the cage overnight. Next morning the shredding of the material and building of a nest-like construct was evaluated on a 5-point scale as follows: 0 = nothing done (> 90% of the material untouched) 1 = paper partially shredded (50–90% remaining intact) 2 = paper mostly torn up (50–90%) but no identifiable nest is built, pieces are outspread 3 = an identifiable but flat nest: > 90% of the paper is torn up, nest walls are higher than mouse height but in less than 50% of the nest circumference 4 = complete nest, > 90% of paper is torn up, the nest walls are higher than mouse height on more than 50% of the nest circumference
2.2. Behavioral protocols 2.2.1. Clasping To score the clasping reflex, mice were kept approximately 1.5 cm above their tail base for about 10 s. Clasping of the hind limbs was scored for each limb separately (0 = hind limb folded backwards and distant from body, angle of limb with body axis > 0°; 0.5 = any phenotype in between 0 and 1; 1 hind limb partially retracted during more than 50% of the time; 3 = limb completely retracted during more than 50% of the time). 2.2.2. Rotarod The Rotarod test was performed to assess motor coordination and balance deficits. Mice were tested in an automated Rotarod device (Model: LE-8200, Five Station Accelerating Rotarod, Bioseb, France), with a maximum of five mice at a time placed on a revolving rod with opaque side walls. The first Rotarod sessions at 2.5 months of age was performed on four consecutive days (familiarization on day 1, four trainings sessions at a constant speed of 10 rpm for a maximum time of 120 s on day 2, four consecutive test trials at 10 rpm for 120 s on day 3, and two acceleration tests wherein the speed of the rod increased from 4 to 40 rpm within 120 s on day 4). At 3 and 4 months of age, the familiarization was omitted. From 4.5 months onwards the training sessions were omitted as well. During the test trials, the mean latency to fall off the Rotarod was recorded and used in the analysis. For the acceleration test, the mean maximum speed and the mean latency to fall off the rod were recorded and analyzed.
2.3. Histopathology Mice were deeply anesthetized with a pentobarbital/chroralhydrate cocktail (60 mg/kg each) and perfused transcardially with ice-cold 0.9% saline followed by 4% paraformaldehyde. After 4 h of postfixation, brains were incubated overnight at +4 °C in 30% sucrose solution and finally stored in a cryoprotectant at −20 °C before sectioning on a freezing microtome into 35 μm thick coronal sections. To visualize pathological tau aggregates, sections were stained for monoclonal anti-human PHF-Tau, clone AT8 (1:1000, Thermo Scientific, IL, USA). After citrate solution pretreatment, sections were treated with 2% hydrogen peroxidase in methanol. After rinsing sections were blocked in 3% bovine serum albumin in TBS-T for 60 min. The primary antibody was also diluted into 3% BSA in TBS-T. The sections were incubated overnight at 4 °C on a shaker table. Following incubation, the sections were rinsed thoroughly with TBS-T and transferred to a solution containing the secondary antibody, goat–anti mouse labelled with Alexa Fluor 488, 1:400 (Invitrogen, Life Technologies
2.2.3. Grid hanging This test was used to determine static muscle force. The mouse was placed on a 20 cm x 25 cm wire grid (grid unit 1 cm by 1 cm) that was 2
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Corporation, OR, USA). After 2-h incubation the sections were rinsed three times and embedded on slides and mounted with Vectashield (Vector Hard Set –mounting medium with DAPI, Vector Laboratories, CA, USA). The sections were photographed using Axio Imager M2 microscope (Zeiss, Jena, Germany) with fluorescent AxioCam MR3 camera attached. The pictures were then examined with Zen pro 2011 blue edition (Zeiss, Jena, Germany).
2.4. Statistics The statistical analyses were done using either Graph Pad Prism 5 or IBM SPSS Statistics 21 software. Analysis of variance for repeated measures (ANOVA-RM) was done for the initial genotype comparison, with the age of the mice as the within-subject and the genotype as the between-subject factor. However, for clasping we used one-sample ttest, since none of the wild-type mice showed the clasping reflex. In addition, since some parameters of the Catwalk gait analysis, such as the paw print areas are dependent on the body weight, we also ran the ANOVA-RM with the body weight as a covariant for these parameters. Depending on the parameter distributions, the post-hoc tests were done with the Least Significant Difference (LSD) test, one-sample t-test or nonparametric Mann-Whitney test. The level of statistical significance was set at p < 0.05.
Fig. 2. Clasping score development between hTau.P301S (hTau) and wild-type (WT) mice during ages 3–6 months. *p < 0.05, **p < 0.01, ***p < 0.001, one-sample t-test. The error bars denote SEMs.
3.3. Impaired muscle coordination and static force in hTau.P301S mice at 4 months of age To assess motor coordination and balance, we used the common Rotarod test. Due to hyperactivity at a young age, hTau.P301S mice needed an extended period of adaptation to the Rotarod test (which was also applied to WT controls) but adapted to repeated testing without further training. We tested the mice in two versions of the test, with a fixed and accelerating speed of rotation. There was an overall genotype × age interaction in both modes (fixed speed: F5,22 = 17.1, p < 0.001; accelerating speed: F5,22 = 27.8, p < 0.001), such that hTau.P301S mice showed a steady decline from 3 months on, reaching significance at 4 months (Fig. 3AB). To assess plain static force with little demand on muscle coordination, we used the grid hanging test for a separate group of young mice. In this test, none of the mice fell before the 5 min cut-off time at 2.5 months of age, but a significant decline appeared in hTau.P301S mice at 3.5 months (Fig. 3C).
3. Results 3.1. Body weight declines in hTau.P301S mice after 4 months of age There was an overall genotype difference in the body weight (F1,18 = 28.5, p < 0.001) and an age × genotype interaction (F38,684 = 82.6, p < 0.001). However, a closer look at the development of body weight suggested that 4 months of age is a turn point for the hTau.P301S mice (Fig. 1). Before the age of 4 months, the body weights of hTau.P301S mice did not differ significantly from WT mice (F1,18 = 3.2, p = 0.09), but showed a decline thereafter while a steady growth continued in WT mice.
3.4. Gait analysis reveals altered posture in hTau.P301S mice as early as at 2.5 months of age To assess the earliest signs of motor impairment, we used the automated Catwalk gait analysis in a separate group of young mice from 2.5 to 4.5 months of age. Fig. 4 shows an example of the automatically assigned footprints. Across these ages, hTau.P301S mice showed an altered posture. The print area of forepaws was significantly smaller in hTau.P301S mice than in WT mice (F1,17 = 22.8, p < 0.001; Fig. 5A), while there was no genotype difference in hindpaws (F1,17 = 1.4, p = 0.25; Fig. 5B). The genotype difference in the forepaw print area remained significant even after correction for the body weight (F1,16 = 7.4, p = 0.015). Further, the base of support (distance between the paws) for forepaws was wider (Fig. 5C) and that of hindpaws narrower in hTau.P301S mice compared to WT mice (Fig. 5D). The difference could be seen already at 2.5 months of age, but clearly accentuated after 3.5 months, which resulted in a significant genotype × age interaction (forepaws: F2,32 = 15.5, p < 0.001; hindpaws: F2,32 = 13.4, p < 0.001). The genotype difference in the base of support of the forepaws disappeared after weight correction (F1,15 = 2.3, p = 0.15) but the genotype × age interaction remained significant (F2,30 = 9.0, p = 0.001). In contrast, both the genotype difference (F1,15 = 10.5, p = 0.006) and the genotype × age interaction (F2,14 = 6.0, p = 0.013) remained significant for the hindpaws after the weight correction. These observations indicate that hTau.P301S mice placed their weight more on hindlimbs. Difficulties in maintaining the balance was also reflected in the swing time of the limbs during the step cycle. The swing time of forelimbs as % of the step cycle showed gradual shortening in hTau.P301S mice as they aged (Fig. 5E), resulting in
3.2. Clasping of the hindpaws appears in hTau.P301S mice after 4 months of age Clasping of the hindpaw was not present in any of the WT mice. However, the clasping score showed a steady increase in hTau.P301S mice from 4 months of age onwards until reaching an asymptote after 5 months (Fig. 2).
Fig. 1. Body weight gain from 2.5 to 6 months of age in hTau.P301S (hTau) and wild-type (WT) mice. Note the clear deviation of the growth curves of the genotypes around 4 months of age. *p < 0.05, **p < 0.01, ***p < 0.001, post-hoc LSD-test for triplet age points. The error bars denote SEMs. 3
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Fig. 3. Tests for balance/dynamic muscle force and static muscle force. (A) Rotarod test with fixed speed (10 rpm). (B) Rotarod test with acceleration. (C) Grid hanging test. The mean ± SEM latency to fall is shown. *p < 0.05, **p < 0.01, ***p < 0.001; A and B, post-hoc LSD-test, C Mann-Whitney test.
5.5 months of age (Fig. 6A). This trend led to a significant genotype × age interaction in the ambulatory distance moved in a fixed time in the test environment (F3,42 = 13.1, p < 0.001). However, the time spend in rearing, which is another indicator of exploratory activity but requires much more balancing, was lower in hTau.P301S than WT mice at all studied ages (genotype main effect: F1,14 = 10.8, p = 0.005; interaction. F3,42 = 0.9, p = 0.47; Fig. 6B).
3.6. Robust impairment in nest building of hTau.P301S mice at all studied ages Finally, we wanted to assess the ability of hTau.P301S mice to perform a complex motor sequence. For such we chose nest building, which is part of the natural behavioral repertoire of mice and does not need additional training. It was obvious that the nest scores remained substantially lower in hTau.P301S mice than their WT controls across all ages from 2.5 to 5.5 months (F1,14 = 138.3, p < 0.001; Fig. 6C). Fig. 4. Example of footprints on the Catwalk apparatus. (Top) A 2.5-month-old wild-type mouse, (bottom) a hTau.P301S transgenic mouse of the same age. LH = left hindpaw, LF = left forepaw, RH = right hindpaw, RF = right forepaw. The hindpaw prints are marked with a red rectangle.
3.7. Earliest signs of pTau aggregation in the brain stem are seen in pontine nuclei To determine a structure-function relationship between the tau pathology and motor impairment as a function of age, we examined the location of pathological hTau aggregation in the brain with the common AT8 antibody recognizing the phosphorylation of serine 202 and threonine 205 of hTau. The spinal cord pathology in the hTau.P301S mouse model has been described earlier [2]. The basal ganglia or cerebellar hemispheres showed no significant AT8 positivity at any age. In contrast, we found substantial aggregation of AT8 positive neurons in the deep cerebellar and medial vestibular nuclei in the pons as early as at 3 months of age (Fig. 7A). There was also scattered AT8 positivity in the lateral vestibular, pontine reticular and gigantocellular nuclei. At 5.5 months, the entire brainstem was full of AT8 positive aggregates (Fig. 7B).
a significant genotype × age interaction (F2,16 = 14.9, p < 0.001), while similar shortening of the hindlimb swing time was present in hTau.P301S mice across ages (Fig. 5F) (genotype main effect: F2,16 = 18.5, p < 0.001; genotype × age interaction: F2,16 = 1.3, p = 0.30). 3.5. Bidirectional change in spontaneous locomotor activity of hTau.P301S mice across ages Young hTau.P301S mice were hyperactive compared to their WT controls. However, as their motor abilities started to decline, their activity level first ostensibly ‘normalized’, followed by hypolocomotion at 4
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Fig. 5. Catwalk gait analysis of hTau.P301S (hTau) and wild-type (WT) mice. (A) Footprint area of forepaws and (B) hindpaws. (C) Base of support (the distance between the supporting pair of paws) for forepaws and (D) hindpaws. (E) Duration of swing (%) of the step cycle for forepaws and (F) hindpaws. *p < 0.05, **p < 0.01, ***p < 0.001; A and B, post-hoc LSD-test. The error bars denote SEMs. Fig. 6. Spontaneous exploratory activity of hTau.P301S (hTau) and wild-type (WT) mice in the TruScan measured with infrared photobeam detectors. (A) Ambulatory distance (gross horizontal locomotion), (B) rearing time. (C) Nest building score on the scale 0 – 4. *p < 0.05, **p < 0.01, ***p < 0.001; A and B, post-hoc LSD-test, C, Mann-Whitney test. The error bars denote SEMs.
5
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
Fig. 7. (A) P-tau aggregations in female hTau.P301S mice at 3 months of age. LEC = lateral entorhinal cortex; PrC = perirhinal cortex; DN = dentate, IpN = interposed, FgN = fastigial cerebelar nucleus; mVe, spVe = medial, spinal vestibular nucleus; PnC = pontine reticular nucleus, caudal part; GiN = gigantocellular nucleus. (B). P-tau aggregations in female hTau.P301S mice at 5.5 months of age. mEC = medial entorhinal cortex,PSub = parasubiculum; dlPAG, laPAG = dorsolateral, lateral periaqueductal gray matter; InCo = intercollicular nucleus; CnF = cuneiform n. ; PnO = pontine reticular n., oral part; PnC = pontine reticular n., caudal part; m5 = motor trigeminal n. ; sp5 = spinal trigeminal n.; 7 = facial nucleus. The coordinates above the sections are from bregma. Scale bar =1.0 mm.
4. Discussion
The early hyperactivity appears to be specific for the hTau.P301S mouse line. Consistent with present findings, a previous study on the early behavioral phenotype of this mouse line also documented hyperlocomotion at 2 months of age [4]. Hyperlocomotion was also observed in the other common mouse line expressing P301S hTau under the mouse prion promoter (aka PS19 line, [6]), but it showed progression with age and became significant only at the age of 7 months [7]. Further, a recently described third P301S hTau mouse (line Tau 58/2; [8]) showed no hyperlocomotion before 12 months of age, albeit axonal pathology and p-tau positive neurons could be observed throughout the brain as early as 3 months of age and impaired Rotarod performance at 6 months [9]. Thus, it appears that the hyperactivity is not directly related to accumulation of hyperphosphorylated tau in the brain of these mouse lines, although hyperlocomotion related to overexpression of mutated human amyloid precursor protein in mice could be reversed by deletion of mouse endogenous tau in one study [10]. The age of 4 months appears to be a watershed for the phenotype of the hTau.P301S mouse. There was a significant decline in body weight
To our knowledge, this is the first systematic study on the development of motor deficits in the hTau.P301S transgenic mouse across the adult life span. We observed two patterns of deficits with different progression with age. Already at the earliest ages studied, 2–3 months, the mice showed marked hyperactivity, abnormal placing of weight on the hindlimbs and defective nest building in their home cage. These behavioral impairments showed no progression with age. The other pattern consisted of progressive development of hindlimb clasping, inability to stay on a rotating rod or hang on a metal grid, and gait impairment. Depending on the measured output parameter, the progressive motor impairment became significant from 3 to 4 months onwards and showed a steep linear trend until the age of 5.5 months with little inter-individual variation. The progressive motor impairment was paralleled by a robust increase in AT8 p-tau positive neurons in deep cerebellar nuclei and pontine brainstem between 3 and 5.5 months of age. 6
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
or hindpaws. Of note, hindlimb paralysis appeared in this mouse at 6 months of age as in the hTau.P301S mouse. We observed slowing of hindlimb swing speed as well, but not before 5.5 months (data now shown). However, the parameters sensitive to body posture (print area and base of support of fore- and hindpaws) and prolongation of the stand phase of the step cycle were abnormal already at earlier ages. We recently published a Catwalk analysis of a mouse with selective degeneration of cerebellar Purkinje cells and secondary lesions in the deep cerebellar nuclei [16]. This mouse showed progressive shortening of stride length and slowing of swing time of both limbs, as well as widening of the forepaw base of support, thus showing partially overlapping pattern of both hTau mouse models. This finding thus lends support to the notion that dysfunction of deep cerebellar nuclei contribute to the observed gait deficit. To conclude, the hTau.P301S mouse demonstrates a fast and steadily progressive motor impairment between 3 and 5.5 months of age. The impairment can be easily monitored by the hindpaw clasping response and a small battery of standard motor tests. During this time window, the motor impairment is paralleled by robust development of tau pathology in deep cerebellar nuclei and pontine brainstem, which is easy to localize in coronal brain sections around the 4th ventricle. This mouse is thus well suited for preclinical studies aiming at slowing down tau pathology before the symptomatic phase.
at 4 months, which is also the age when many indicators of motor impairment show a significant change. The time course of decline in the body weight and motor functions raise the question whether the motor impairment may result from, for instance, decreased muscle mass. We did not measure the muscle mass directly but found no indication of atrophy of special muscle groups. The most direct measure of static muscle force in our test battery was the grid hanging task where impairment of hTau P301S mice was evident as early as 3.5 months of age, i.e.two weeks before the body weight started to decline (Fig. 3C). Another parameter sensitive to muscle force was explorative rearing. Indeed, the rearing time showed a decline in hTauP301S mice between 3.5 and 4.5 months of age, but for some reason a similar decline was also seen in WT mice (Fig. 6B). On the other hand, the clasping reflex appeared in hTau P301S mice precisely after 4 months of age but it is totally independent of muscle force. Taken together, muscle weakness per se cannot explain the observed motor phenotype. The abnormal posture of hTau.P301S mice, with increased weight on hindlimbs and wider base of support of the forelimbs, may be explained by early tau pathology in lumbal spinal motoneurons. However, the abnormal posture was present already at 2.5 months of age when there was no evidence of hindlimb weakness in terms of static muscle force as assessed in the grid hanging test. Furthermore, the relative shortening of the swing during the step cycle in hTau.P301S mice in fact means that the swing speed was increased. This fits poorly with dysfunction at the spinal level. Another, and more plausible, explanation for the abnormal posture is early tau pathology in the brainstem. Namely, both lateral vestibular nucleus and pontine reticular formation send projections to the lumbal spinal motoneurons and thereby influence the muscle tone of the hindlimbs and body posture [11]. Recent evidence suggests that stimulation of the ventral part of the mesencephalic locomotor center in the pontine reticular formation induces rearing in rats, while it lesioning results in atonia [12]. Impairment of this brainstem neuron group that maintains the tonus of the hindlimb extensors may well account for the abnormal posture and decreased rearing of hTau.P301S mice despite hyperactivity, which usually involves rearing besides horizontal locomotion. The most striking sign of abnormality in hTau.P301S mice was clasping of hindpaws when the mouse was lifted by the tail. It started consistently in all mice around 4 months of age and reached it full extent by 5 months of age. Similar hindpaw clasping was reported in the PS19 mouse starting from 3 months of age [6]. The clasping response has been reported in several genetically modified mouse lines, but most consistently in mice with primary pathology in the cerebellum or striatum [13]. Since the hTau.P301S mouse did not express tau pathology in the striatum, and deep cerebellar nuclei constitute the only output channel of cerebellum, the observed pathological tau aggregations in fastigial and interposite cerebellar nuclei are likely candidates to account for the abnormal hindpaw clasping. Further, the rapid progression of intracellular p-tau aggregation in this brain region matches the appearance of the clasping response. In agreement with the report on early behavioral abnormalities hTau.P301S mice [4] we found significant impairment in the Rotarod performance at 4 months of age. Since successful performance in this task requires both motor coordination and balance, it is sensitive to a variety of lesions. Typically, genetic mouse models of cerebellar ataxia display impairment in the Rotarod test [14]. In addition, the vestibular nuclei play a key role in maintaining the balance. Thus, the observation of early p-tau deposits in the fastigial cerebellar nucleus and vestibular nuclei at 3 months, and their rapid progression thereafter, likely contribute to this deficit. The Catwalk gait analysis has been employed in only one study on a tauopathy mouse so far [15]. A mouse expressing truncated human tau and tau aggregation mainly in the brainstem showed no significant deviance from the controls at 3 and 4 months of age, but at 5 months the stride length became shorter and swing time slower; however, the body posture was not changed, as evidenced by print areas of the fore-
Declaration of Competing Interest BR and EY and SC are employees of reMYND N.V. Acknowledgement The study was supported by EU-Horizon 2020 project PANA (grant 686009-2). References [1] A.D. Sperfeld, M.B. Collatz, H. Baier, M. Palmbach, A. Storch, J. Schwarz, K. Tatsch, S. Reske, M. Joosse, P. Heutink, A.C. Ludolph, FTDP-17: an early-onset phenotype with parkinsonism and epileptic seizures caused by a novel mutation, Ann. Neurol. 46 (Nov (5)) (1999) 708–715. [2] B. Allen, E. Ingram, M. Takao, M.J. Smith, R. Jakes, K. Virdee, H. Yoshida, M. Holzer, M. Craxton, P.C. Emson, C. Atzori, A. Migheli, R.A. Crowther, B. Ghetti, M.G. Spillantini, M. Goedert, Abundant tau filaments and nonapoptotic neurodegeneration in transgenic mice expressing human P301S tau protein, J. Neurosci. 22 (2002) 9340–9351. [3] X. Chai, S. Wu, T.K. Murray, R. Kinley, C.V. Cella, H. Sims, N. Buckner, J. Hanmer, P. Davies, M.J. O’Neill, M.L. Hutton, M. Citron, Passive immunization with anti-Tau antibodies in two transgenic models: reduction of Tau pathology and delay of disease progression, J. Biol. Chem. 286 (Sep (39)) (2011) 34457–34467. [4] M.L. Scattoni, L. Gasparini, E. Alleva, M. Goedert, G. Calamandrei, M.G. Spillantini, Early behavioural markers of disease in P301S tau transgenic mice, Behav. Brain Res. 208 (Mar (1)) (2010) 250–257. [5] R.M. Deacon, Assessing nest building in mice, Nat. Protoc. 1 (3) (2006) 1117–1119. [6] Y. Yoshiyama, M. Higuchi, B. Zhang, S.M. Huang, N. Iwata, T.C. Saido, J. Maeda, T. Suhara, J.Q. Trojanowski, V.M. Lee, Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model, Neuron 53 (Feb (3)) (2007) 337–351. [7] M. Gerges, M. Dumont, C. Stack, C. Elipenahli, S. Jainuddin, N.N. Starkova, L. Yang, A.A. Starkov, F. Beal, Behavioral deficit, oxidative stress, and mitochondrial dysfunction precede tau pathology in P301S transgenic mice, FASEB J. 25 (2011) 4063–4072. [8] J. van Eersel, C.H. Stevens, M. Przybyla, A. Gladbach, K. Stefanoska, C.K. Chan, W.Y. Ong, J.R. Hodges, G.T. Sutherland, J.J. Kril, D. Abramowski, M. Staufenbiel, G.M. Halliday, L.M. Ittner, Early-onset axonal pathology in a novel P301S-Tau transgenic mouse model of frontotemporal lobar degeneration, Neuropathol. Appl. Neurobiol. 41 (2015) 906–925. [9] M. Kassiou, M. Przybyla, C.H. Stevens, J. van der Hoven, A. Harasta, M. Bi, A. Ittner, A. van Hummel, J.R. Hodges, O. Piguet, T. Karl, G.D. Housley, Y.D. Ke, L.M. Ittner, J. Eersel, Disinhibition-like behavior in a P301S mutant tau transgenic mouse model of frontotemporal dementia, Neurosci. Lett. 631 (2016) 24–29. [10] M. Michikawa, M. Yoshikawa, Y. Soeda, O.F.X. Almeida, A. Takashima, Tau depletion in APP transgenic mice attenuates task-related hyperactivation of the Hippocampus and differentially influences locomotor activity and spatial memory, Front. Neurosci. 12 (2018) 124. [11] T. Terashima, K. Inoue, Y. Inoue, K. Mikoshiba, Y. Tsukada, Observations on the brainstem-spinal descending systems of normal and reeler mutant mice by the
7
Behavioural Brain Research 376 (2019) 112158
H. Koivisto, et al.
[15] I. Zimova, V. Brezovakova, T. Hromadka, P. Weisova, V. Cubinkova, B. Valachova, P. Filipcik, S. Jadhav, T. Smolek, M. Novak, N. Zilka, Human truncated tau induces mature neurofibrillary pathology in a mouse model of human tauopathy, J. Alzheimers Dis. 54 (2016) 831–843. [16] R.R. Nair, H. Koivisto, K. Jokivarsi, I.J. Miinalainen, K.J. Autio, A. Manninen, P. Poutiainen, H. Tanila, J.K. Hiltunen, A.J. Kastaniotis, Impaired mitochondrial fatty acid synthesis leads to neurodegeneration in mice, J. Neurosci. 38 (2018) 9781–9800.
retrograde HRP method, J. Comp. Neurol. 225 (1984) 95–104. [12] D. Sherman, P.M. Fuller, J. Marcus, J. Yu, P. Zhang, N.L. Chamberlin, C.B. Saper, J. Lu, Anatomical location of the mesencephalic locomotor region and its possible role in locomotion, posture, cataplexy, and parkinsonism, Front. Neurol. 6 (2015) 140. [13] R. Lalonde, C. Strazielle, Brain regions and genes affecting limb-clasping responses, Brain Res. Rev. 67 (2011) 252–259. [14] J.N. Crawley, What’s Wrong with My Mouse? 2nd ed., John Wiley & Sons, Hoboken, NJ, USA, 2007, pp. 80–81.
8