Skill learning induced plasticity of motor cortical representations is time and age-dependent

Abstract

Movement representations in the motor cortex can reorganize to support motor skill learning during young adulthood. However, little is known about how motor representations change during aging or whether their change is influenced by continued practice of a skill after it is learned. We used intracortical microstimulation to characterize the organization of the forelimb motor cortex in young and aged C57/BL6 mice after short (2–4 weeks) or long (8 weeks) durations of training on a skilled reaching task or control procedures. In young mice, a short duration of reach training increased the area of proximal forelimb movement representations at the expense of distal representations. Following a longer training duration, ratios of proximal to distal movements returned to baseline, even with ongoing practice and skill maintenance. However, lingering changes were evident in thresholds for eliciting distal forelimb movements, which declined over the longer training period. In aged mice, movement representations and movement thresholds failed to change after either duration of training. Furthermore, there was an age-related loss of digit representations and performance decrements on other sensorimotor tests. Nevertheless, in quantitative measures of reaching success, aged mice learned and performed the skilled reaching task at least as well as younger mice. These results indicate that experience-driven topographical reorganization of motor cortex varies with age, as well as time, and is partially dissociable from behavioral performance. They also support an enduring capacity to learn new manual skills during aging, even as more youthful forms of cortical plasticity and sensorimotor function are lost.

Introduction

Age-related decreases in manual dexterity (Carmeli et al., 2003, Smith et al., 1999), compounded by decreases in speed and strength (Ranganathan, Siemionow, Sahgal, & Yue, 2001) may result in an increased risk of injury and trouble performing tasks of daily living (Incel, Sezgin, As, Cimen, & Sahin, 2009). This contributes to a loss of independence and may also be an early predictor of cognitive decline (Kluger, Gianutsos, Golomb, Ferris, & Reisberg, 1997). However, in healthy subjects, motor training can attenuate the loss of dexterity associated with aging. Training on a skilled finger movement task improved the ability of older adults to control pinch force, hand steadiness, and manual speed, compared to untrained older adults (Ranganathan, Siemionow, Sahgal, Liu, & Yue, 2001). A better understanding of healthy aging effects on forelimb motor function and motor cortical organization could lead to the development of strategies to extend youth-like motor ability and manual dexterity into old age.

Age-related changes in motor performance are associated with functional alterations in several brain regions. Transcranial magnetic stimulation (TMS) and functional magnetic resonance imaging (fMRI) studies indicate greater activation of premotor, supplementary motor, sensory and cognitive areas in the aged brain during motor task performance (Heuninckx et al., 2005, Heuninckx et al., 2008, Hutchinson et al., 2002, Talelli et al., 2008, Ward and Frackowiak, 2003). The increase in cognitive and sensory activation is thought to reflect an increased reliance on sensory cues and more cognitive control during movement, which may allow older adults to produce motor behavior that is more similar to that seen in young adulthood (Heuninckx et al., 2005, Heuninckx et al., 2008).

Rats and mice provide useful models for studying age-related changes in manual skills due to their high degree of forelimb dexterity (Whishaw and Coles, 1996, Whishaw et al., 1998). In the somatosensory cortex, the topographical complexity of the forelimb representation of rats degrades as they age (Coq & Xerri, 2000) and this coincides with impairments in forelimb and hindlimb walking behavior (David-Jürgens et al., 2008, Spengler et al., 1995). The somatosensory map maintains some complexity (Coq & Xerri, 2001) and walking behavior more closely resembles that of young animals (David-Jürgens et al., 2008) when rats are raised in enriched environments beginning immediately after weaning. In contrast to the somatosensory cortex, little is known about how the organization and plasticity of the rodent motor cortex change during aging.

In young adult mice, motor cortical dendrites add new spines within the first hour after training on a novel skilled reaching task. The total spine density returns to baseline when a skill has been practiced for ∼2 weeks, but there is selective survival of new spines formed during early skill acquisition (Xu et al., 2009). In young adult rats, Kleim, Barbay, and Nudo (1998) found that training on a skilled reaching task increases representation areas for movements necessary for task performance at the expense of representations of lesser-used movements. Molina-Luna, Hertler, Buitrago, and Luft (2008) demonstrated that the areal extent of the forelimb movement representations in rats increases during acquisition of a motor skill and returns to baseline after 8 days without training, but the ability to perform the task is maintained. It is unknown why the map returns to baseline without a loss of motor skill: whether this results from the cessation of practice or is reflective of a typical stage of motor cortical reorganization that occurs even with continued practice. Studies of humans with extensive motor skill practice show conflicting results on this question as well. For example, several studies have shown that professional musicians (Elbert et al., 1995, Gaser and Schlaug, 2003), professional badminton players (Pearce, Thickbroom, Byrnes, & Mastaglia, 2000), and blind Braille readers (Pascual-Leone et al., 1993) have a larger area of motor cortex corresponding to the trained hand compared to novices. However, Pascual-Leone, Wassermann, Sadato, and Hallett (1995) showed a decrease in motor representations in Braille readers after a weekend without practice, and others have shown that professional musicians show less activation in motor areas compared to novices (Jäncke et al., 2000, Krings et al., 2000). Additionally, two recent studies have shown that the auditory receptive fields of rats undergo expansion at early stages of training on an operant conditioning task (Takahashi et al., 2011), a tone discrimination task, or after nucleus basalis stimulation with tone pairing (Reed et al., 2011) followed by shrinkage in later stages of training. Thus, the characteristics and dynamics of cortical map plasticity, especially as they relate to long-term training, are complex and insufficiently understood. More information is needed on the characteristics of motor cortical plasticity, even in young adult brains, to understand how changes in these characteristics contribute to motor behavior across the lifespan.

The main purpose of the current study was to investigate age-related changes in the baseline organization of motor cortex and its response to motor skill learning and practice. An additional goal was to determine the effect of extended practice on the maintenance of map reorganization after it has occurred. Young and aged mice were trained on one of two behavioral tasks, the Pasta Matrix Reaching Task or a dexterous food (pasta) handling task, or served as untrained controls. To determine the effect of training duration on the organization of motor cortex, young and aged mice underwent intracortical microstimulation (ICMS) procedures after either a short duration (2–4 weeks) or a long duration (8 weeks) of training. Our results indicate that the organization and plasticity of the forelimb motor cortex is altered in the aged brain, but the ability to learn a skilled motor task is maintained. Additionally, in young mice, the reorganization of movement representations during skill learning is transient even with ongoing skill practice.