Research report
Time course of choice reaction time deficits in the HdhQ92 knock-in mouse model of Huntington's disease in the operant Serial Implicit Learning Task (SILT)

https://doi.org/10.1016/j.bbr.2008.01.020Get rights and content

Abstract

A range of transgenic and knock-in mouse models of Huntington's disease have been created since identification in 1993 of the disease mutation in the HD gene. Knock-in models that express the full-length mutant protein tend to exhibit less severe behavioural deficits than transgenic models and so require more sensitive tasks in order to reveal impairments. To achieve this, we therefore used a Serial Implicit Learning Task (SILT), which measures serial reaction times to visual stimuli, requiring detection and responding in both predictable and unpredictable locations in the 9-hole operant chamber. We have previously reported that knock-in HdhQ92/Q92 mice exhibit a modest impairment in learning the SILT tasks at 4 months of age, prior to the formation of overt neuronal nuclear inclusions. In the present study we have explored the time course of the development of impairments from 5 to 14 months of age. The deficit previously found in accuracy and reaction time was present at all ages examined in these HdhQ92/Q92 mice; the deficit was not progressive, and did not correlate with the evolution of neuronal nuclear inclusions.

Introduction

Huntington's disease (HD) is an inherited neurological disorder that causes a progressive loss of motor control, cognitive deficits, and emotional disturbance [3]. The disease is the result of a single gene mutation involving an expanded CAG repeat within the HD gene encoding the protein huntingtin on chromosome four [37]. The production of mutant huntingtin results in selective neuronal loss that starts and is most marked in the striatum [42], but involves progressive atrophy to include other areas of the basal ganglia and cerebral cortex at later stages of the disease [11].

A number of genetic models of the disease have now been generated, to include both transgenic [13], [24], [29], [32], [34], [41], [48] and knock-in [22], [23], [26], [33], [43], [45] models. Although these models all express an expanded CAG repeat, the spatial and temporal development of the pathology in these different lines varies greatly. Those knock-in models that express full-length mutant huntingtin tend to have a less severe phenotype than the transgenic models carrying truncated forms of the human gene [25]. Largely for reasons of simplicity and efficiency, most studies exploring novel therapeutics have addressed the severe and rapid onset phenotypes in the latter transgenic strains [4]. By contrast, the more subtle full-length mutant huntingtin knock-in models have been relatively neglected, but may provide a valid model of the time course and phenotypes of the human disease. Deficits in these animals need to be uncovered with sensitive behavioural tests, so these more genetically precise models can be used to test potential therapies. Moreover, detailed analysis of the cognitive dysfunctions in these animals may identify deficits that link these lines more closely to HD.

One of the many cognitive abilities affected in HD is visuomotor function, the ability to co-ordinate motor responses to visual information with both speed and accuracy. For example, HD patients are impaired on simple and choice reaction time tasks, which require the subjects to respond rapidly to stimuli presented in different locations on a screen [16], [18], [35], [47]. In one version of this task, repeated sequences of stimuli are embedded within the random sequence, revealing an additional impairment of HD patients to exhibit ‘implicit learning’ [18]. In order to access similar aspects of visuomotor learning and performance in rodents, we have developed the Serial Implicit Learning Task (SILT) in which the rats or mice must learn to respond rapidly to lights presented within a spatial array; whereas most stimuli occur in a random sequence, specific combinations always occur in a repeated sequence. In previous studies we have shown that rats and mice can learn the SILT and that performance is disrupted by bilateral striatal lesions[17], [39]. Full-length mutant htt knock-in mice of the HdhQ92/Q92 strain have a modest but significant impairment in learning the task at 4 months of age [40]. As HD is progressive in nature, in the present study we explore the temporal development of this impairment, examining a group of HdhQ92 in the SILT at 2–3-month intervals from 5 to 14 months of age.

Section snippets

Subjects

HdhQ92/Q92 mice on a mixed 129SvEv/CD1 background strain (founders from JAX, Maine, USA) were crossed in house onto C57BL/6J (Harlan, UK), for four generations. Heterozygous Hdh+/Q92 mice were mated and the homozygous HdhQ92/Q92 offspring and their wild-type littermates used in the behavioural experiments. Nine male (4 Hdh+/+, 5 HdhQ92/Q92) mice and 12 female (6 Hdh+/+, 6 HdhQ92/Q92) mice were initially housed in pairs, in standard cages at the start of the training period. However, all the

Results

Both the HdhQ92/Q92 and Hdh+/+ mice were able to learn the ‘Continuous’ version of the implicit learning task (i.e. with S2 illuminated until the animal responded) at 4–5 months of age, similar to previous description at 4 months of age [40]. Once performance had stabilized at the 5-month time point, the number of trials the HdhQ92/Q92 mice initiated, appeared to be fewer than their Hdh+/+ littermates. However, this did not reach significance (218.4 ± 20.5 vs. 282.4 ± 24.0 per day, genotype: F1,19 = 

Discussion

As has been previously demonstrated [40], both the wild-type and HdhQ92/Q92 mice were able to perform the SILT and make use of the concealed predictable information. A complex motor learning deficit was present at all ages examined, as previously seen in 4-month-old HdhQ92/Q92 mice [40]. This deficit became more pronounced with the shortening of the S2 stimulus to 0.5 s (increasing the attentional demands of the task [for review Ref. [30]]).

The HdhQ92/Q92 mice were slower than their Hdh+/+

Acknowledgements

This work was funded by the BBSRC and MRC. We thank Prof Gillian Bates for the generous gift of the S830 antibody and also Jane Heath and Polyanas Cheng for their technical assistance.

References (48)

  • G. Bates et al.

    Huntington's disease

    (2002)
  • M.F. Beal et al.

    Experimental therapeutics in transgenic mouse models of Huntington's disease

    Nat Rev Neurosci

    (2004)
  • J.C. Bensadoun et al.

    Free operant and discrete trial performance of mice in the nine-hole box apparatus: validation using amphetamine and scopolamine

    Psychopharmacology (Berl)

    (2004)
  • D. Craufurd et al.

    Behavioral changes in Huntington disease

    Neuropsychiatry Neuropsychol Behav Neurol

    (2001)
  • C. Exner et al.

    Implicit and explicit memory after focal thalamic lesions

    Neurology

    (2001)
  • J.D. Gabrieli et al.

    Intact mirror-tracing and impaired rotary-pursuit skill learning in patients with Huntington's disease: evidence for dissociable memory systems in skill learning

    Neuropsychology

    (1997)
  • N. Georgiou et al.

    The Simon effect and attention deficits in Gilles de la Tourette's syndrome and Huntington's disease

    Brain

    (1995)
  • A. Ho et al.

    NEST-HD consortium, profile of cognitive progression in early Huntington's disease

    Neurology

    (2003)
  • T. Humby et al.

    Visuospatial attentional functioning in mice: interactions between cholinergic manipulations and genotype

    Eur J Neurosci

    (1999)
  • A.R. Isles et al.

    Measuring impulsivity in mice using a novel operant delayed reinforcement task: effects of behavioural manipulations and d-amphetamine

    Psychopharmacology (Berl)

    (2003)
  • M. Jahanshahi et al.

    A comparative study of simple and choice reaction time in Parkinson's, Huntington's and cerebellar disease

    J Neurol Neurosurg Psychiatry

    (1993)
  • J.R. Jay et al.

    An operant serial implicit learning task (SILT) in rats: task acquisition, performance and the effects of striatal lesions

    J Neurosci Methods

    (2007)
  • B.J. Knowlton et al.

    A neostriatal habit learning system in humans

    Science

    (1996)
  • A.D. Lawrence et al.

    Executive and mnemonic functions in early Huntington's disease

    Brain

    (1996)
  • Cited by (32)

    • Using Drosophila models of Huntington's disease as a translatable tool

      2016, Journal of Neuroscience Methods
      Citation Excerpt :

      These mice have been created to express the full-length HTT gene containing various lengths of polyQ repeats, namely the Q92, Q111 and Q150 (Brooks et al., 2012a; Menalled et al., 2003; Trueman et al., 2008; Wheeler et al., 2000). These show both pathological hallmarks and behavioral traits, recapitulating the human condition (Bayram-Weston et al., 2012a,d; Brooks et al., 2012a; Menalled et al., 2003; Trueman et al., 2008, 2012; Wheeler et al., 2000). However, due to their subtle phenotypes, and minimal loss of MSNs most research has been conducted using homozygous KI mice.

    • Analysis of Nonmotor Features in Murine Models of Huntington Disease

      2015, Movement Disorders: Genetics and Models: Second Edition
    • Cognitive deficits in animal models of basal ganglia disorders

      2013, Brain Research Bulletin
      Citation Excerpt :

      This B–D predictable combination is the implicit learning probe, performance of which is measured against the anti-sense (D–B) combination. In a series of studies on in the HdhQ92 mice, we have reported that homozygotes are impaired both on the S1 and S2 choice accuracy and had increased reaction times to the stimuli by 4 months of age [170], but that task performance remained stable with age [171]. In neither study was an implicit learning deficit found, with both wildtype and HdhQ92 mice being able to benefit from the predictable stimuli [170,171].

    • Light and electron microscopic characterization of the evolution of cellular pathology in Hdh <sup>Q92</sup> Huntington's disease knock-in mice

      2012, Brain Research Bulletin
      Citation Excerpt :

      The HdhQ92 knock-in mouse model has ∼90 CAG repeats inserted into exon 1 of the mouse HD gene [60]. This mouse line demonstrates mild motor-learning impairments from 4 months of age [6,50–52] but otherwise demonstrates little behavioural pathology until the mice are in the latter stages of life. As part of an ongoing programme to provide a comparative characterisation of different HD mouse lines [3–5], the present study evaluated the nature and the spatial and temporal distribution of neuropathology in the HdhQ92 mouse line.

    • A novel extended sequence learning task (ESLeT) for rodents: Validation and the effects of amphetamine, scopolamine and striatal lesions

      2012, Brain Research Bulletin
      Citation Excerpt :

      Within our programme seeking to develop novel neuroprotective and cell therapies for basal ganglia disorders, we have sought to develop an operant version of implicit learning within a serial reaction time task for rodents. Using the ‘9-hole box’ operant apparatus [27], the serial implicit learning task (SILT) was the first implementation of this strategy [31], and reveals clear deficits in speed and latencies of serial reaction time responses in rats and mice associated with striatal lesions [4,15,31] or mutations of the huntingtin gene [3,32,33], modelling the human HD pathology. However, as in PD patients, both lesioned and mutant mice continue to exhibit the specific benefit of implicit knowledge, as indicated by higher performance on predictable than unpredictable trials, even though their overall performance was markedly impaired on both the accuracy and latency measures in comparison to normal controls [3,4,31–33].

    • Proteomic changes in the brains of Huntington's disease mouse models reflect pathology and implicate mitochondrial changes

      2012, Brain Research Bulletin
      Citation Excerpt :

      This indicates the existence of widespread molecular changes from at least 10 months. Though few gene expression changes were observed at 3 months in this line [45], this may be consistent with behavioural studies showing a behavioural impairment at 4 months, which was present through to 14 months without being progressive, despite the appearance of neuronal nuclear inclusions during this period [69,70]. The 18 month HdhQ150 showed a proportion of differentially expressed proteins nearly double that in the 18 month HdhQ92.

    View all citing articles on Scopus
    View full text