, J. David Jentsch2 and Marie-FranÇoise Chesselet1
(1)
Department of Neurology, The David Geffen School of Medicine at UCLA, Los Angeles, CA, USA
(2)
Department of Psychology, The David Geffen School of Medicine at UCLA, CA, 90095-1769, Los Angeles
Abstract
Parkinson’s disease (PD) is primarily recognized as a motor disorder; however, patients also present with a wide range of nonmotor manifestations. Cognitive dysfunctions in nondemented PD patients can occur early in the disease and primarily consist of deficits in executive function. Because it can be assessed with noninvasive measurement tools, cognitive dysfunction could be evaluated to determine the effects of potential disease-modifying agents in patients. A challenge is to reproduce these deficits in animals for preclinical drug testing. Genetic mouse models of PD have been generated based on mutations causing rare familial forms of PD. Although only a few models show extensive nigrostriatal dopamine cell loss, several present extensive anomalies in functions that are also altered in premanifest phases of PD. Here we review the few studies that have so far investigated cognitive function in these new models.
Key words
Alpha-synucleinReversal learningParkinPrefrontal cortexLocus coeruleus1 Introduction
Parkinson’s disease (PD) is primarily characterized by motor impairments including tremor, rigidity, bradykinesia, and postural instability that are associated with the progressive loss of dopamine neurons in the substantia nigra pars compacta. However, in addition to the motor symptoms, patients also develop a wide variety of nonmotor symptoms such as cognitive dysfunction, autonomic dysfunction, sleep disorders, and olfactory impairments indicating PD is a systemic disorder. Both environmental and genetic factors have been linked to PD and suggest that sporadic PD may arise from an interaction between environmental toxins combined with a genetic susceptibility.
2 Cognitive Dysfunction in Early PD
The nonmotor symptoms associated with PD are eliciting growing interest because many of these symptoms may appear early in the disease process and greatly contribute to the overall quality of life of patients with the disorder. Among these nonmotor symptoms, cognitive dysfunction includes subtle impairments in implicit memory, attention, visuospatial skills, and executive function (1, 2). The estimated prevalence of cognitive dysfunction in PD ranges from 20% to 40% (3–5). Several studies show that PD patients, as a group, exhibit impairments in executive function similar to patients with frontal lobe lesions (6), including increased perseverative errors in the Wisconsin Card Sorting Task (7), attention deficits such as ignoring irrelevant stimuli properties (8), and attentional set shifting (9). Patients also have shown impaired habit learning (1). Deficits in executive function and habit learning have been associated with alterations in the striatum (1, 10), and dysfunction within the prefrontal cortex (11, 12). Dopamine agonists can have both beneficial and detrimental effects on cognitive function in patients and in healthy volunteers (13–16), suggesting that extranigral pathology may contribute, at least in part, to the early cognitive dysfunction in PD. Indeed, Braak and colleagues proposed the concept of progressive pathology in PD through analysis of α-synuclein pathology in the brains of PD patients at varying stages of the disease and of individuals without overt neurological disorders (17). These authors have shown that early α-synuclein pathology begins in subcortical nuclei including the dorsal motor nucleus X, olfactory regions, and in the locus coeruleus prior to the development of α-synuclein pathology in the substantia nigra that leads to manifest PD (17). The potential dysfunction of the locus coeruleus in the early stages of PD has important implications because the locus coeruleus has diffuse projections to cortical areas, including the prefrontal cortex. At later stages, progressive pathology to the cerebral cortex itself may underlie the worsening of cognitive dysfunction that is observed in some patients with advanced PD.
3 Cognitive Dysfunction in Toxin Models of Parkinsonism
Traditional models of PD are based on the use of neurotoxins, including 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and 6-hydroxydopamine, to kill nigrostriatal dopaminergic neurons. Cognitive dysfunction has been detected in some of these toxin-induced models of Parkinsonism. Depending on the dosing regimen, MPTP-treated nonhuman primates show deficits in a range of tasks that require working memory and/or response control/inhibition (18–21). Similarly, MPTP-treated and 6-hydroxydopamine-treated mice show cognitive impairments in alternation, habituation, and spatial memory (22–24). Although these models have been important for our understanding of cognitive deficits associated with nigrostriatal dopamine cell loss, they lack the broad extranigral pathology observed in PD.
4 Cognitive Dysfunction in Genetic Models of Parkinsonism
Several types of genetic mouse models of PD have been generated. Some models are based on the deletion or inactivation of factors that are important for the differentiation and/or maintenance of nigrostriatal dopaminergic neurons such as Nurr1, Pitx3, and engrailed (25–27). These models exhibit progressive loss of nigrostriatal dopaminergic neurons and show motor and/or affective deficits (25, 28, 29), but none has yet been examined for cognitive dysfunction. More relevant to PD are models based on mutations known to cause rare familial forms of PD. Few lead to overt, progressive nigrostriatal dopaminergic cell loss in the lifetime of the mouse, but they have strong construct validity because they are based on mechanisms known to cause PD in humans. Furthermore, they provide the opportunity to study the early nonmotor symptoms associated with PD (30).
A variety of PD-causing mutations has now been expressed in several different lines of mice (30, 31). Of particular interest are mice that overexpress the presynaptic protein α-synuclein. Mutations and multiplication of the α-synuclein gene are sufficient to induce PD (32–35). In sporadic PD, α-synuclein is a major component of Lewy bodies, the pathological hallmark of PD (36). Alpha-synuclein accumulates in selective neuronal populations throughout the body (17), indicating that it is involved in sporadic as well as familial forms of PD.
4.1 Cognitive Deficits in Mice Overexpressing Alpha-Synuclein
Several laboratories have begun to study cognitive function in different lines of α-synuclein transgenic mice (Table 1). One line of mice that expresses the human A30P α-synuclein mutation and presents with α-synuclein inclusions predominantly in the brain stem and spinal cord develops motor impairments after the first 12 months of life primarily due to motor neuron pathology (37). However, these mice also have α-synuclein inclusions in the amygdala that may be associated with neuropsychiatric impairments such as cognitive dysfunction and anxiety. Indeed, when transgenic and wild-type mice were tested for visuospatial learning and memory in the Morris water maze, for fear conditioning, and for active avoidance at 4 and 12 months of age prior to the onset of motor impairments, the transgenic mice displayed age-dependent alterations in all three tasks suggesting cognitive dysfunction and increased anxiety (38) (Table 1).
Table 1
Cognitive impairments in genetic mouse models of Parkinsonism
α-synuclein transgenic mice | Cognitive impairment |
|---|---|
Thy1-aSyn (50) | ↓ Cumulative accuracy in reversal learning task |
Thy1-A30P (38) | ↓ Preference for platform quadrant in Morris water maze |
CaM_α-syn (39) | ↓ Retention in Morris water maze |
Parkin knockout mice | Cognitive impairment |
Exon 3 deleted (54) | ↓ Spontaneous alternation in T-maze |
Exon 3 deleted (53) | Impaired spatial learning in Morris water maze |
In a novel conditional mouse model generated to express high levels of human wild-type α-synuclein in midbrain and forebrain regions under the prion promoter, α-synuclein pathology developed in both the substantia nigra and hippocampus leading to motor and cognitive impairments (39). In the Morris water maze, transgenics did not differ from controls in swim speed, acquisition, or in the 24 h probe trial. However, they showed reduced retention following a 7-day probe trial. Although the cognitive deficit appears to be associated with hippocampal dysfunction, which is not prominent in PD patients, it does highlight the relationship between α-synuclein pathology and cognitive deficits (Table 1).
We have extensively characterized a line of mice with broad overexpression of human wild-type α-synuclein under the Thy1 promoter (Thy1-aSyn), exhibiting proteinase K-resistant α-synuclein aggregates in areas similar to those affected in PD, including the locus coeruleus and substantia nigra (40, 41, Hutson et al. in preparation). These mice do not show overt cell loss up to 8 months of age, but exhibit progressive loss of tyrosine hydroxylase-containing terminals in the striatum (Mortazavi et al., in preparation). Thy1-aSyn mice have alterations in striatal dopamine synapses including a chronic increase in basal levels of extracellular dopamine in the striatum (42), abnormal behavioral responses to dopaminergic agonists (43), and abnormal electrophysiological responses to dopaminergic agonists and antagonists in striatal slices (44). In addition, Thy1-aSyn mice show olfactory, cardiovascular, and gastrointestinal dysfunction, all nonmotor symptoms associated with early PD (45–47). Recent data indicates that noradrenaline in the prefrontal cortex is decreased in Thy1-aSyn mice (Hean et al. 2010, unpublished observations), and given the critical role of noradrenaline in modulating aspects of cognition (48), these observations suggest that Thy1-aSyn mice may have impairments in cognitive function.
Because patients show difficulty in strategy switching, we tested Thy1-aSyn and wild-type mice in a reversal learning task that assesses cognitive flexibility (49). Male Thy1-aSyn and wild-type littermates at 4–5 months of age were trained on an operant serial reversal task. Mice were trained to respond into one of two illuminated apertures to obtain food reward and the contingencies were reversed after performance criteria were achieved. In this test we found that Thy1-aSyn mice can learn a simple operant strategy as well as controls, but show greater difficulty in their ability to switch their response at reversal, compared to wild-type mice, although they were able to eventually reach criteria and learn the reversed contingency (50).
Since dopamine agonists can have both beneficial and detrimental effects on cognition (13–15), we assessed the effect of two different doses of l-dopa (15 and 25 mg/kg, i.p.) in Thy1-aSyn and wild-type mice in the reversal learning paradigm (50). In wild-type mice the low dose of l-dopa had no effect on accuracy, while the higher dose reduced accuracy (Table 2). In contrast, both doses had a positive effect on reversal accuracy in the Thy1-aSyn mice. Although l-dopa is known for its affects on the dopamine system, it also increases noradrenaline and it is possible that improvement in the l-dopa studies may be due, in part, to stimulation of noradrenaline receptors, since Thy1-aSyn mice have decreased noradrenaline in the prefrontal cortex (Hean et al. 2010, unpublished observation). To test this hypothesis, we assessed the effect of two different doses of the α2-noradrenergic agonist guanfacine (0.1 and 1.0 mg/kg, i.p.) on the reversal task. Similar to l-dopa, both doses of guanfacine had a positive effect on reversal accuracy in the Thy1-aSyn mice, while the low dose had no effect and the higher dose reduced accuracy in wild-type mice. This indicates that Thy1-aSyn mice have cognitive impairments that can be improved pharmacologically (Table 2).
Table 2
Effect of l-dopa and guanfacine in Thy1-aSyn and wild-type mice in a reversal task
Cumulative accuracy | ||
|---|---|---|
Drug (compared to saline) | Wild-type | Thy1-aSyn |
L-DOPA (15 mg/kg, i.p.) | No effect | ↑ |
L-DOPA (25 mg/kg, i.p.) | ↓ < div class='tao-gold-member'>
Only gold members can continue reading. Log In or Register to continue
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
| |