Elsevier

Behavioural Brain Research

Volume 311, 15 September 2016, Pages 340-353
Behavioural Brain Research

Research report
High-speed video gait analysis reveals early and characteristic locomotor phenotypes in mouse models of neurodegenerative movement disorders

https://doi.org/10.1016/j.bbr.2016.04.044Get rights and content

Highlights

  • Full body high-speed video gait analysis increases specificity in motor testing.

  • Comprehensive locomotor profiles are generated by a custom-built algorithm.

  • Gait patterns of ALS, Huntington and cerebellar ataxia models differ substantially.

  • Early and specific gait changes were found for each of the three models analyzed.

  • Analysis of locomotor function is essential for the development of new therapies.

Abstract

Neurodegenerative diseases of the central nervous system frequently affect the locomotor system resulting in impaired movement and gait. In this study we performed a whole-body high-speed video gait analysis in three different mouse lines of neurodegenerative movement disorders to investigate the motor phenotype. Based on precise computerized motion tracking of all relevant joints and the tail, a custom-developed algorithm generated individual and comprehensive locomotor profiles consisting of 164 spatial and temporal parameters. Gait changes observed in the three models corresponded closely to the classical clinical symptoms described in these disorders: Muscle atrophy due to motor neuron loss in SOD1 G93A transgenic mice led to gait characterized by changes in hind-limb movement and positioning. In contrast, locomotion in huntingtin N171-82Q mice modeling Huntington’s disease with basal ganglia damage was defined by hyperkinetic limb movements and rigidity of the trunk. Harlequin mutant mice modeling cerebellar degeneration showed gait instability and extensive changes in limb positioning. Moreover, model specific gait parameters were identified and were shown to be more sensitive than conventional motor tests. Altogether, this technique provides new opportunities to decipher underlying disease mechanisms and test novel therapeutic approaches.

Introduction

Neurodegenerative disorders of the central nervous system, depending on the structures involved, often affect movement and gait. In amyotrophic lateral sclerosis (ALS) muscle atrophy, paralysis and spasticity, due to spinal cord and motor cortex pathology, result in gait disturbances [1]. Damage to basal ganglia structures in Huntington’s disease (HD) alternatively causes a hyperkinetic disorder (chorea) combined with a loss of voluntary movements (bradykinesia and rigidity) [2]. Disorders primarily affecting cerebellar structures, such as spinocerebellar ataxias, result in impaired balance and coordination (ataxic gait) [3].

To better understand the underlying pathophysiological mechanisms, several mouse models representing these neurodegenerative disorders have been established: (1) SOD1 G93A transgenic (tg) mice demonstrate ALS-like motor neuron loss and hind-limb paralysis [4]; (2) Huntingtin N171-82Q mice develop abnormal gait and behavioral abnormalities [5]; (3) Harlequin mutants bear a mutation leading to cerebellar degeneration and ataxic syndrome [6].

Comprehensive analysis of motor and gait function is a cornerstone for the pre-clinical development of new therapies. Assessment techniques based on behavioral tests such as rotarod, grip strength or scoring systems are characterized by several limitations, including a restriction to a few specific aspects of locomotion, non-physiologic test conditions and a dependency on users and motivational factors. They often fail to fully capture delayed onset following treatment intervention (sensitivity) and progressive motor dysfunction (qualitative and quantitative changes). Therefore, more sophisticated techniques were needed.

Gait analysis is promising for the evaluation of motor deficits, as gait is a fundamental, physiological and unforced form of locomotion with direct clinical relevance. However, available systems focusing on ventral plane video gait analysis produce results that vary widely, fail to reproduce or even contradict previous findings [7], [8], [9], [10], [11], [12]. This is not surprising, as many characteristic changes in limb positioning and movement dynamics in ALS and HD models can only be observed in the lateral plane. Lately, innovative systems including machine learning algorithms (NeuroCube®) and analysis of the lateral view (MotoRater, Locomouse) became available and were used to examine gait more thoroughly [13], [14], [15], [16], [17]. But the maximum potential of lateral plane videography has not yet been exploited, as analyses were restricted to single functional aspects of gait and a limited number of gait parameters in few models.

In this study we performed a comprehensive whole-body locomotion and gait analysis of the mentioned mouse models of neurodegenerative movement disorders using the MotoRater (TSE-Systems, Bad Homburg, Germany), a high-speed video tracking system [18]. This system films the spontaneous gait from three sides (ventral and lateral planes) to track bony landmarks.

We followed an exploratory approach (using 17 tracking points) to study the movement of limbs, trunk and tail in order to create a highly comprehensive locomotor profile and developed a custom-built algorithm, which converts and interprets the raw tracking data and calculates 164 objective spatial and spatiotemporal parameters describing locomotion. Based on this extensive dataset, we identified a small set of relevant tracking points and parameters for each model, which specifically describe gait disturbances, and may be used in following studies to assess potential therapeutic approaches with a higher throughput. Further, we examined effects of locomotion speed and body weight, as previous gait analysis studies pointed out their influence on many individual locomotor parameters [19], [20], [21], [16]. Moreover, we compared the three models representing different human neurodegenerative gait disorders and observed specific and distinct motor profiles. Several gait changes observed in the three models correspond closely to the classical clinical symptoms described in these disorders.

Lastly, a higher sensitivity of this system compared to conventional motor tests was proven, revealing substantial locomotor deficits detectable prior to the onset of clinical impairments on conventional tasks.

Section snippets

Animals

All animal experiments were approved by the veterinarian authorities of the Canton of Zurich (animal license numbers: 109/2010, 183/2012, 229/2013), in compliance with Swiss law (“455.163 Tierversuchsverordnung” 2010) and international guidelines for care and use of animals.

All animals were held and housed 3–5 per cage under specific pathogen-free conditions in an air-conditioned and temperature controlled environment (22 °C ± 1 °C). 3 weeks after birth, tg animals were identified by standard PCR

Comprehensive locomotor profile describes gait in great detail

Our custom-built algorithm allowed the calculation of a comprehensive locomotor profile. This profile consists of 164 objective spatial and spatiotemporal parameters describing movement of every tracking point. For a detailed representation of all computed parameters see Supplementary Table S1. We identified a set of 25 parameters, of which each reflects specific gait deficiencies in one or more models, shown in Table 1. Parameters were classified into following three functional groups: (1)

Discussion

The examination of unforced locomotion has direct clinical relevance, as it is fundamental and physiologic. Unlike ventral plane approaches, whole-body video gait analysis is not restricted to paw placements and coordination, but provides a global assessment of locomotion and motor abilities. Consequently, our MotoRater analysis identified specific gait disturbances for each of the three different mouse models analyzed.

Mainly altered hind-limb movement, accompanied by some changes in

Conflict of interest

The authors have no conflicts of interest to disclose.

Acknowledgements

The authors would like to thank Anna Jeske, Belinda Ries and Stefan Preisig for excellent technical assistance and contributions to the project as well as Kristina Hersch for useful discussion and comments.

References (49)

  • R. Mancuso et al.

    Evolution of gait abnormalities in SOD1G93A transgenic mice

    Brain Res.

    (2011)
  • P.O. Fernagut et al.

    A simple method to measure stride length as an index of nigrostriatal dysfunction in mice

    J. Neurosci. Methods

    (2002)
  • M. Ooms et al.

    Early decrease of type 1 cannabinoid receptor binding and phosphodiesterase 10A activity in vivo in R6/2Huntington mice

    Neurobiol. Aging

    (2014)
  • P.N. Pallier et al.

    The detection and measurement of locomotor deficits in a transgenic mouse model of Huntington’s disease are task- and protocol-dependent: influence of non-motor factors on locomotor function

    Brain Res. Bull.

    (2009)
  • C.R. Hayworth et al.

    Pre-symptomatic detection of chronic motor deficits and genotype prediction in congenic B6. SOD1G93A ALS mouse model

    Neuroscience

    (2009)
  • J.D. Rothstein

    Current hypotheses for the underlying biology of amyotrophic lateral sclerosis

    Ann. Neurol.

    (2009)
  • C.A. Ross et al.

    Huntington disease: natural history, biomarkers and prospects for therapeutics

    Nat. Rev. Neurol.

    (2014)
  • A.H. Ropper et al.

    Chapter 7. Disorders of stance and gait

    Adams Victors Princ. Neurol.

    (2014)
  • M.E. Gurney et al.

    Motor neuron degeneration in mice that express a human Cu, Zn superoxide dismutase mutation

    Science

    (1994)
  • G. Schilling et al.

    Intranuclear inclusions and neuritic aggregates in transgenic mice expressing a mutant N-terminal fragment of huntingtin

    Hum. Mol. Genet.

    (1999)
  • J.A. Klein et al.

    The Harlequin mouse mutation downregulates apoptosis-inducing factor

    Nature

    (2002)
  • C.M. Wooley et al.

    Gait analysis detects early changes in transgenic SOD1(G93A) mice

    Muscle Nerve

    (2005)
  • T.S. Guillot et al.

    Treadmill gait analysis does not detect motor deficits in animal models of Parkinson’s disease or amyotrophic lateral sclerosis

    J. Mot. Behav.

    (2008)
  • T.G. Hampton et al.

    Treadmill gait analysis characterizes gait alterations in Parkinson’s disease and amyotrophic lateral sclerosis mouse models

    J. Mot. Behav.

    (2009)
  • Cited by (28)

    • Stride-level analysis of mouse open field behavior using deep-learning-based pose estimation

      2022, Cell Reports
      Citation Excerpt :

      These studies did not adjust for body size difference or in some cases for stride speed. We tested SOD1-G93A transgenes and appropriate controls at 64 and 100 days, during time of disease onset (Wooley et al., 2005; Preisig et al., 2016; Vergouts et al., 2015; Mancuso et al., 2011; Knippenberg et al., 2010). We do not see significant differences in body length or weight (Figures S3C and S11), but changes are seen in stride speed (Figure S3B).

    • A simple method to study motor and non-motor behaviors in adult zebrafish

      2019, Journal of Neuroscience Methods
      Citation Excerpt :

      Various paradigms of behavior such as cognitive behavior, social interactions and locomotor activity had been analysed manually in different animal models to correlate the behavioural patterns with physiological findings of specific disorder (Ali et al., 2011; Piato et al., 2011; Bartholomew et al., 2015; Peng et al., 2016; Franco-Restrepo et al., 2019). Analysis of motor functions has great clinical relevance and can be best achieved by analysing the locomotor activity (Preisig et al., 2016). Locomotion analysis is an efficient, non-invasive, economical approach and accelerating technology with broad range of applications spanning drug discovery and screening, detection of neurobehavioral disorders and chemical toxicology (Tang and Su, 2013).

    • Mouse models of central nervous system ageing

      2017, Drug Discovery Today: Disease Models
    View all citing articles on Scopus
    View full text