Animal Models of Huntington's Disease

Animal Models of Huntington’s Disease

Shilpa Ramaswamy, Jodi L. McBride, and Jeffrey H. Kordower

Abstract Huntington’s disease (HD) is a neurological disorder caused by a genetic mutation in the IT15 gene. Progressive cell death in the striatum and cortex, and accompanying declines in cognitive, motor, and psychiatric functions, are characteristic of the disease. Animal models of HD have provided insight into disease pathology and the outcomes of therapeutic strategies. Earlier studies of HD most often used toxin-induced models to study mitochondrial impairment and excitotoxicity-induced cell death, which are both mechanisms of degeneration seen in the HD brain. These models, based on 3-nitropropionic acid and quinolinic acid, respectively, are still often used in HD studies. The discovery in 1993 of the huntingtin mutation led to the creation of newer models that incorporate a similar genetic defect. These models, which include transgenic and knock-in rodents, are more representative of the HD progression and pathology. An even more recent model that uses a viral vector to encode the gene mutation in specific areas of the brain may be useful in nonhuman primates, as it is difficult to produce genetic models in these species. This article examines the aforementioned models and describes their use in HD research, including aspects of the creation, delivery, pathology, and tested therapies for each model. Key Words: 3-nitropropionic acid; animal models; Huntington’s disease; knock-in mice; quinolinic acid; transgenic rodents

H

untington’s disease (HD1) is an invariably fatal neurodegenerative disorder that results from a mutation in the interesting transcript 15 (IT151) gene (The Huntington’s Disease Collaborative Research Group 1993). The mutation is an expanded trinucleotide repeat (CAG) in

Shilpa Ramaswamy, M.S., is a Ph.D. candidate and Jeffrey H. Kordower, Ph.D, is the Director of the Section of Neurobiology, the Director of the Research Center for Brain Repair and Jean Schweppe Armour Professor of Neurological Sciences in the Department of Neuroscience at Rush University Medical Center in Chicago. Jodi L. McBride, Ph.D., is a postdoctoral fellow in the Department of Internal Medicine at the University of Iowa, Iowa City. Address correspondence and reprint requests to Dr. Jeffrey H. Kordower, Department of Neuroscience, Rush University Medical Center, 1735 W. Harrison Street, Suite 300, Chicago, IL 60612 or email [email protected].

356

exon 1 of the huntingtin (htt1) gene, leading to the production of mutant htt protein with an abnormally long polyglutamine repeat. The mutated protein aggregates in striatal neurons as well as neurons in other regions of the neuraxis, such as the cortex, thalamus, hypothalamus, and the substantia nigra pars compacta. Interestingly, htt-containing aggregates do not usually form in the globus pallidus, hippocampus, or cerebellum, despite cell loss in these areas. Mutant htt has a toxic gain of function that causes cell death via mechanisms that remain unclear. Clinically, the most obvious symptoms of HD involve involuntary hyperkinetic (choreaform) movements of the arms, legs, and face. But the severe cognitive and personality changes are the most devastating to HD patients and most troubling for their caregivers. Researchers are using animal models of HD to study the disease pathogenesis, to elucidate areas of the brain involved in structural and functional decline, and to evaluate potential therapeutic interventions. The most reliable models of HD recapitulate the neuropathology and symptomology of the disease. Pathologically, HD primarily affects the striatum and cerebral cortex, as is evident from the abundant inclusion formation and cell loss in the medium-sized spiny neurons (MSNs1) of the striatum that express gamma-aminobutyric acid (GABA1) as their neurotransmitter and in the large glutamatergic pyramidal neurons of the cortex. There are two populations of GABAergic striatal neurons and neuronal death does not occur simultaneously in them (Figure 1; Reiner et al. 1988). In early stages of the disease, neurons that coexpress enkephalin and project to the external segment of the globus pallidus via the indirect pathway are preferentially vulnerable. In normal basal ganglia circuitry, the indirect pathway is involved in the inhibition of neurons in the motor cortex and hence the inhibition of voluntary movements (Albin et al. 1989; Alexander and Crutcher 1990). Death of these enkephalin-containing medium spiny neurons causes the hallmark hyperkinetic, choreaform (dance-like) movements typical of HD. In late stages of the disease, death occurs in

1 Abbreviations used in this article: 3-NP, 3-nitropropionic acid; AAV, adeno-associated viral; ATP, adenosine triphosphate; CNTF, ciliary neurotrophic factor; GABA, ␥-aminobutyric acid; GDNF, glial-derived neurotrophic factor; HD, Huntington’s disease; htt, huntingtin; IT15, interesting transcript 15; KA, kainic acid; MA, malonic acid; mRNA, messenger RNA; MSN, medium spiny neuron; NMDA, N-methyl-Daspartic acid; QA, quinolinic acid; ROS, reactive oxidative species; TCA, tricarboxylic acid cycle; YAC, yeast artificial chromosome

ILAR Journal

Figure 1 Aberrant basal ganglia circuitry in early and late-stage HD. (A) Early in HD, GABAergic neurons in the striatum that project to the external segment of the globus pallidus degenerate (dashed arrow). The resulting inactivation of the subthalamic nucleus (thick arrow) leads to disinhibition of the thalamus and increased stimulation of the premotor and supplementary motor cortices (thick arrow). (B) Later in the course of the disease, GABAergic neurons in the striatum that project to the internal segment of the globus pallidus degenerate (dashed arrow). This loss reduces the inhibition on GPi neurons and increases their firing (thick arrow), which results in an enhanced inhibition of the thalamus and decreased stimulation of the premotor and supplementary motor cortices (thin arrow).

medium spiny neurons that coexpress substance P and project to the globus pallidus via the direct circuit. This pathway is normally involved in the initiation of voluntary movements (Albin et al. 1989; Alexander and Crutcher 1990), and so the death of these GABAergic/substance P neurons produces hypokinetic and parkinsonian symptoms. During progression of the disease, neurons in the caudate die earlier and more extensively than those in the putamen (Vonsattel et al. 1985). Degeneration of neurons in the striatum does not occur uniformly as there is a selective preservation of interneurons. Large and medium aspiny interneurons that express acetylcholine (Ferrante et al. 1987a), NADPH-diaphorase (Ferrante et al. 1987b), and somatostatin (Dawbarn et al. 1985) are well preserved. Aspiny striatal interneurons that coexpress neuropeptide Y are upregulated in some HD brains (Dawbarn et al. 1985). HD spares calbindin- and calretinin-expressing large interneurons, but GABAergic MSNs that coexpress calbindin are preferentially lost (Cicchetti et al. 2000). Cortical degeneration is also a prominent feature of HD pathology and contributes, at least in part, to the deficits in higher-order cognitive function. It is progressive, heterogeneous, and localized in select areas of the cortex. Cell loss occurs in several layers of the cortex but is most prominent in layers V and VI (Hedreen et al. 1991). Hedreen and colleagues have reported a 71% neuronal loss in layer V and 57% loss in layer VI of the dorsal frontal cortex (Hedreen et al. 1991). Selemon and colleagues reported an overall 25% decrease in thickness of the dorsolateral prefrontal cortex, Volume 48, Number 4

2007

with an increase in glial cell density in layer VI (Selemon et al. 2004). They also showed a 16% decrease in neurons in layer III, a 31% decrease in layer V, and a 37% decrease in layer VI in Broadman’s area 9. Because layer VI of the cortex projects to the thalamus, claustrum, and other cortical areas, but not to the striatum, it is likely that cortical cell loss is a primary process and not secondary to wallerian degeneration of axon terminals due to striatal cell loss. Animal models of HD principally mimic symptoms that result from pathology seen in the striatum and its connections, but it is important to create animal models that also represent, to the extent possible, the genetic and molecular mechanisms that underlie degenerative processes in the human disease. To that end, the combination of knowledge about the genetic basis of this disease and the emergence of transgenic and gene transfer technologies has permitted the creation of animal models of HD that share the same genetic defect as the human disease. Because the main cause of HD is a mutation in the htt gene, most current models include the insertion of this mutation (or a portion thereof) in the rodent genome. The expanded polyglutamine stretch in mutant htt confers a toxic gain of function on the protein and thus leads to functional deficits and specific pathologies that mimic the human form of the disease in these genetically altered animals. The most widely studied animal models employ rodents and, more recently, nonhuman primates, but animal models of HD exist in nonmammalian species as well. Invertebrate models like Caenorhabditis elegans and Drosophila mela357

nogaster allow for rapid and high-throughput testing of specific hypotheses and novel therapeutic strategies. The C. elegans model expresses expanded polyglutamine repeats in the worm nervous system (Brignull et al. 2006). Long CAG repeats result in the formation of polyglutamine aggregates in muscle cells and overall decreased motility. The drosophila model of HD expresses expanded polyglutamine repeats in the eye of the fly (Jackson et al. 1998). In this model, although the eye develops normally, photoreceptors progressively develop inclusions and subsequently degenerate. Invertebrates’ short life spans make it easy and inexpensive to generate large numbers of animals. However, thorough evaluation of disease processes and research to develop novel therapeutics ultimately require models with more complexity. Most animal models of HD fall into two broad categories, genetic and nongenetic. Historically, nongenetic models have dominated the field of HD research. Although George Huntington first described HD in 1872, researchers did not identify the actual genetic mutation responsible for the disease until 1993, which delayed the development of appropriate genetic models until the last decade. Nongenetic models typically induce cell death either by excitotoxic mechanisms or by disruption of mitochondrial machinery (Table 1). Quinolinic acid (QA1) and kainic acid (KA1)

have been the two most commonly used excitotoxic agents in both rodent and primate models of HD. These amino acids (QA and KA) induce cell death by binding to their cognate receptors, N-methyl-D-aspartic acid (NMDA1) and non-NMDA, respectively, on striatal neurons. The mitochondrial toxins 3-nitropropionic acid (3-NP1) and malonic acid (MA1) have also been used in both rodents and nonhuman primates to produce cell death in striatal neurons via inhibition of the Complex II (succinate dehydrogenase) of the tricarboxylic acid cycle and the electron transport chain in mitochondria, effectively reducing production of adenosine triphosphate (ATP1). Emerging molecular technology has enabled the development of genetic murine and, more recently, rat models that attempt to capture the hereditary nature of HD. These models introduce in the rodent’s germline genes that express the mutated htt protein. Newly created mouse and rat lines express either a truncated or full-length form of the mutant htt gene inserted either randomly into the genome (transgenic models) or specifically into the rodent htt gene locus (knock-in models). The R6/1 and R6/2 transgenic mouse models were first characterized by Bates and colleagues (Mangiarini et al. 1996) and are still the most widely used transgenic mouse models. These mice express mutant exon 1 of the human htt gene and exhibit both early

Table 1 Toxin models Animal model Quinolinic acid (QA)

3-Nitropropionic acid (3-NP)

358

Species Rat (Sprague-Dawley and Fischer) (Bordelon et al. 1997; Ribeiro et al. 2006), mouse (McLin et al. 2006) and nonhuman primate (Emerich et al. 2006; Kendall et al. 2000) Rat (all except Fischer) (Ouary et al. 2000), mouse (Yang et al. 2005), nonhuman primates (Palfi et al 2000; Ludolph et al. 1991; Bead et al. 1993)

Mode of Cells administration affected

Method of cell death

Intrastriatal injections, intraputamenal injections (Foster et al. 1984)

Excitotoxicity (Beal et al. 1991; Ferrante et al. 1993; Roberts et al. 1993)

Systemic injections, intrastriatal injections, intraputamenal injections

Enkephalin, substance P, calbindin, parvabumin, aspiny interneurons upregulated (Beal et al. 1986, 1991; Albin et al. 1990) GABAergic medium spiny neurons (Beal et al. 1993), aspiny interneurons, lateral striatal most affected (Blum et al. 2001)

Motor symptoms

Cognitive symptoms

Hyperkinesia, apomorphineinduced dystonia and dyskinesia (Vazey et al. 2006), spontaneous dyskinesia with higher doses (Borlongan et al. 1995)

Visuospatial deficits, procedural memory deficits, poor memory recall (Furtado and Mazurek 1996; Shear et al. 2002)

Mitochondrial Hyperkinesia (low impairment by dose) (Borlongan irreversibly et al. 1997; Palfi inhibiting et al. 1996), succinate hypokinesia dehydrogenase (high dose), (Alston et al. apomorphine1977; Coles induced dystonia et al. 1979) and dyskinesia, spontaneous dyskinesia with long-term administration (Brouillet et al. 1995)

Deficits in ORDT in nonhuman primates (Palfi et al. 1996), radial arm water maze test of working and reference memory in rats, deficits in habituation to open-field apparatus

ILAR Journal

and severe behavioral and anatomical symptoms. R6/1 mice express 114 CAG repeats and R6/2 mice, 150 repeats. N171-82Q transgenic mice created by Borchelt and colleagues express the first 171 amino acids of the htt protein bearing 82 CAG repeats (Schilling et al. 1999). Behavioral and anatomical symptoms in this model develop over a more protracted time course than those of the R6 models. The creation of yeast artificial chromosome (YAC1) transgenic mice involves cloning an artificial yeast vector that contains an expanded polyglutamine repeat into the mouse genome (Hodgson et al. 1999), and the creation of a transgenic rat model involves a similar process (von Horsten et al. 2003). These rats express 51 CAG repeats and display both behavioral and anatomical deficits. Rats, in general, tend to live approximately 1 year longer than mice and have a more complex behavioral repertoire, making the transgenic rat model an attractive candidate for thorough, longterm therapeutic studies. Knock-in models that use gene targeting to insert expanded CAG repeat coding regions into the mouse htt locus have also recently been created. This article reviews each of these models of HD. Recently investigators have also used viral vectors to overexpress the human mutant htt gene in the normal striatum of rats and nonhuman primates, enabling the insertion of mutant htt directly into specific cells of interest. Rats and nonhuman primates that express mutant htt from viral vectors exhibit behavioral and pathological abnormalities similar to those seen in HD (Palfi et al., personal communication). In the absence of a transgenic primate model, this approach is an extremely valuable tool for mechanistic and therapeutic studies in species higher on the phylogenetic scale and capable of higher-order functional processing. In this review we will compare and contrast the different models from mechanistic, pathological, and behavioral perspectives and will briefly discuss how these models can be optimized to more accurately model the behavioral and pathological abnormalities seen in the human form of Huntington’s disease.

Toxin-Induced Models Excitotoxic Models: Quinolinic Acid Initial toxin models that relied on excitotoxic mechanisms utilized kainic acid (KA) (Coyle and Schwarcz 1976). Although these experiments were landmark studies, investigators learned that KA produced remote lesions and, at higher concentrations, destroyed fibers of passage, and so they tested alternative excitotoxins such as ibotenic acid and quinolinic acid (QA). For a variety of reasons detailed below, QA became the preferred excitotoxin for use in HD studies. The first reports of QA, which is one of two products formed from the metabolism of tryptophan via the kynurenine pathway, noted its excretion in the urine of rats that received a diet high in tryptophan (Singal et al. 1946; Henderson and Hirsch 1949). These reports also observed that QA was a common component in the urine of several Volume 48, Number 4

2007

species and that the administration of dietary tryptophan (Henderson et al. 1949) augmented its presence. Tryptophan crosses the blood-brain barrier using transporters shared by other neutral amino acids (Hargreaves and Pardridge 1988). In the brain, tryptophan is taken up by astrocytes, macrophages, microglia, and dendritic cells and converted into kynurenine (Ruddick et al. 2006). In the presence of the enzyme 3-hydroxyanthranilic acid oxygenase, a series of enzymatic reactions converts kynurenine to QA. Normal levels of QA do not cause damage, but only small increases in QA levels cause toxicity. In the postmortem brains of HD victims, Schwarcz and colleagues demonstrated an increase in the enzyme 3-hydroxyanthranilic acid oxygenase, relative to the level in control brains (Schwarcz et al. 1988), with the greatest increase in the striatum, the most vulnerable neurons in the HD brain. These data suggested that in the HD striatum, there is a higher than normal conversion of kynurenine to QA at toxic levels and that this pathway may facilitate neuronal death in the disease. Interestingly, administration of QA to the mouse striatum causes an upregulation of the htt protein (Tatter et al. 1995), linking the expression of this toxin with the critical protein in human HD pathogenesis. Furthermore, studies have shown that overexpressing wild-type htt can protect neuronal cell cultures from QA-induced toxicity (Leavitt et al. 2006). This close association of QA to the htt protein and potentially to the protein-related toxicity makes it a relevant model for testing specific hypotheses related to HD. QA is incapable of crossing the blood-brain barrier and is therefore experimentally administered directly to the striatum (Foster et al. 1984). It causes striatal neurodegeneration in rats (Figure 2A) (Bordelon et al. 1997; Ribeiro et al. 2006), mice (McLin et al. 2006), and primates (Emerich et al. 2006; Kendall et al. 2000) in a pattern similar to that seen in human HD. QA lesions cause symptoms that often mimic deficits seen in early (but not later) stages of HD. For example, the lesions produce hyperactivity in animal models, but the hypoactivity that occurs later in the disease is not modeled by any dose of the toxin. In rodents, unilateral QA lesions cause asymmetric rotational behavior induced by the dopamine agonist apomorphine (Vazey et al. 2006); rotational behavior results from an imbalance in dopamine signaling between the lesioned and intact hemisphere. QA causes degeneration of striatal neurons expressing dopamine receptors and thereby attenuates dopamine signaling. QA-lesioned rats exhibit rotational behavior in the direction ipsilateral to the lesioned hemisphere. To examine similar impairments in motor performance in the absence of drug impairments, a body swing test has been used in QAlesioned rats (Borlongan et al. 1995). When unilaterally lesioned rats are suspended by the tail they preferentially undulate toward the lesioned hemisphere, indicating impairments in the function of the contralateral striatum. Correction of this imbalance via, for example, neural transplants prevents this abnormal undulation (Borlongan et al. 1998). Rats that receive unilateral lesions with QA become impaired in the use of their contralateral forelimb. These rats 359

Figure 2 Neurotoxin models of HD. (A) Coronal section of a rat brain lesioned with QA (150nM). This brain has been stained with the neuronal marker NeuN to indicate pronounced cell loss in the striatum. A unilateral lesion was created by injecting QA directly into one striatum. Significant cell loss appears in the lesioned hemisphere compared to the intact side. (B) Coronal section of a rat brain lesioned with 3-NP (40 mg/kg/day). This brain has been stained with the neuronal marker NeuN. A bilateral lesion was created by injecting the rat subcutaneously for 6 days. Comparable cell loss appears in both striata with predominant cell loss occurring in the lateral striatum (arrows).

preferentially use their ipsilateral forelimb for exploratory and rearing behavior (McBride et al. 2004) QA lesions in rodents produce deficits on cognitive tests that use the Morris water maze, the radial arm water maze, and the T-maze (Furtado and Mazurek 1996; Isacson et al. 1984; Shear et al. 1998). These tests examine working and reference memory function and require intact visuospatial skills and memory recall. In humans disruption of the corticostriatal circuitry impairs these cognitive functions. A similar impairment in QA-lesioned rats indicates that an intact striatum is necessary for cognitive performance on these tests. Indeed, the seminal paper by Isacson and col360

leagues (Isacson et al. 1984), demonstrating that grafts of fetal striatum can reverse T-maze deficits that result from bilateral QA striatal lesions, indicates that specific cognitive features relevant to HD can be modeled by lesions restricted to the striatum and are distinct from deficits engendered by degeneration of cortical circuitry. An advantage to using QA as a model for HD is its easy applicability to more complex animals such as nonhuman primates. Monkeys that receive QA lesions in the putamen, but not the caudate, demonstrate impaired reaching with the contralateral forelimb (Kendall et al. 2000) as well as ipsilateral apomorphine-induced rotations. Unilateral lesions restricted to the posterior putamen cause dystonia and dyskinesias after apomorphine administration (Burns et al. 1995). Bilateral lesions of the posterior putamen in nonhuman primates cause chorea-like symptoms 48 hours after the lesion and thus make this an interesting model for study. Importantly, the QA model influences cognitive function as well. Bilateral lesions of both the caudate and putamen in nonhuman primates cause deficits in the objectretrieval detour task (ORDT) (Roitberg et al. 2002). The ORDT examines procedural memory and frontal control of motor task planning and requires intact frontostriatal circuitry, which is disrupted in the bilateral QA model. The test requires subjects to retrieve a reward from a transparent box with only one open side by circumventing the sides of the box. QA-lesioned monkeys possess the motor ability to perform the task but inappropriately retain the impulse to reach straight ahead for the reward versus correctly searching for and identifying the opening to the box. The increase in barrier hits for the lesioned monkeys, as compared to unlesioned controls, indicates deficits in motor planning. These deficits mirror cognitive changes in human HD that include visuospatial deficits (Mohr et al. 1991), poor memory recall (Butters et al. 1994), and impaired procedural memory (Butters et al. 1985). There are several similarities between pathology observed in the HD brain and in the QA model. As with HD, QA-induced GABAergic cell death involves degeneration of enkephalin-positive and substance P-positive medium spiny neurons (Beal et al. 1986). We note, however, that this pattern appears later in the disease process and is more similar to juvenile HD, which affects both populations of medium spiny neurons equally (Albin et al. 1990). In HD, there is an at least threefold increase in striatal somatostatin and neuropeptide neurons (Beal et al. 1988; Dawbarn et al. 1985), and a similar twofold increase in these populations of cells after QA administration (Beal et al. 1991). QA-treated rats also show a decrease in the neurotransmitter GABA and an increase in the GABAA receptor in the substantia nigra (Nicholson et al. 1995), similar to those seen in HD. Another reason for the widespread use of QA in HD research is that the cell death it causes may mimic the mechanism of neuronal death seen in HD brains (Beal et al. 1991; Ferrante et al. 1993; Roberts et al. 1993). Although the exact mechanism of cell death in HD is unknown, a slow, glutamate-induced excitotoxic cell death has been ILAR Journal

postulated. In the HD brain there is reduced ATP production, leading to a secondary impairment of the Na+-K+ ATPase pump, which normally functions to maintain electronic gradients across cellular membranes during action potentials. In the HD brain, the membrane remains chronically depolarized after the firing of action potentials, allowing for the expulsion of the Mg2+ that normally blocks NMDA receptors. When NMDA receptors remain open for long periods of time there is an influx of Ca2+ into the cell, leading to oxidative damage (Beal 1992). The population of neurons that die in HD, the medium spiny neurons of the striatum, are NMDA receptor rich and are therefore more vulnerable to excitotoxic cell death than other cell types (Gardian and Vecsei 2004). QA works in a similar manner in animal models (Stone and Perkins 1981): administration of QA to the NMDA-receptor-rich striatum causes an increase in Ca2+ influx, a decrease in ATP production, and corresponding excitotoxic cell death in a manner that mimics certain aspects of neurodegeneration in human HD (Bordelon et al. 1997). Although excitotoxicity is closely associated with necrosis, cell death in the QA model also mimics apoptosis seen in the HD brain. Researchers have reported DNA fragmentation and TUNEL staining consistent with apoptosis in early and intermediate-stage HD (Portera-Cailliau et al. 1995). This apoptotic cell death occurs in both neurons and glial cells of the striatum. TUNEL staining also appears in neurons of the striatum after QA administration, but is undetectable 7 days after injection. Thus the type of degeneration seen in the model mimics what occurs in HD although the pace of degeneration in the model is acute and therefore does not accurately model the human disease.

Metabolic Models: 3-Nitropropionic Acid 3-nitropropionic acid (3-NP) is a toxin that irreversibly inhibits the mitochondrial enzyme succinate dehydrogenase (Alston et al. 1977; Coles et al. 1979). The effects of 3-NP in the brain were first discovered when children from China ingested sugarcane infested with Arthrinium (Ming 1995; Liu et al. 1992). The metabolism of this fungus produced 3-NP, which invariably caused cell death in the caudate and putamen and which induced severe dystonia in these children (Ludolph et al. 1991). The 3-NP model is reliable for studying HD because it mimics a downstream process of cell death seen in the HD brain, namely mitochondrial impairment. Impaired glucose metabolism in brain cells as a result of enzyme deficiencies causes decreased production of ATP. Several enzymes involved in the tricarboxylic acid cycle (TCA1) and the electron transport chain are downregulated in the brain of HD patients. Aconitase and complex II, III, and IV activities are reduced in the caudate and putamen of HD brains (Browne et al. 1997; Gu et al. 1996; Tabrizi et al. 1999). Disruption of mitochondrial activity is associated with the abnormal formation of reactive oxidative species (ROS1). Inhibition Volume 48, Number 4

2007

of enzymes in the electron transport chain can lead to an increase in electron leakage from the mitochondria and production of ROS like the superoxide radical (O2−•), hydrogen peroxide (H2O2), and the hydroxyl radical (OH−•). ROS cause damage to cellular membranes and genetic material as demonstrated by an increase in the DNA damage marker 8-hydroxydeoxyguanosine (OH8dG) (Browne et al. 1997). 3-NP causes mitochondrial damage by inhibiting succinate dehydrogenase, an enzyme involved in both the TCA and the electron transport chain (Alston et al. 1977; Coles et al. 1979). Researchers have replicated this finding in rats (Blum et al. 2001), mice (Yang et al. 2005), and nonhuman primates (Beal et al. 1993; Ludolph et al. 1991; Palfi et al. 2000). 3-NP crosses the blood-brain barrier and can be administered systemically to rats, mice, and nonhuman primates. Both the subcutaneous osmotic pump and direct subcutaneous injection are effective for the systemic administration of 3-NP to the rat, but the injection delivers a more precise dose as it is adjusted daily for the animal’s weight. This feature is crucial for 3-NP dosing, as a rat’s weight can decrease by as much as 20 grams per day during 3-NP administration (Ramaswamy and McBride, unpublished observations). Specific rat strains respond very differently to the 3-NP toxin (Ouary et al. 2000). Fischer rats are the most susceptible but display significant variability in response to the toxin, making studies using this species difficult to perform. The Lewis rat is less susceptible to 3-NP, but with appropriate dosing shows consistent lesions and behavioral deficits. The 3-NP model is capable of mimicking both the hyperkinetic and hypokinetic symptoms of HD depending on the time course of administration. When 3-NP is delivered in two individual doses, rats mainly suffer from hyperkinetic symptoms analogous to those seen in early to mid-stage HD (Borlongan et al. 1997). The administration of more than four injections of 3-NP, in contrast, causes hypokinetic movements similar to those that appear in latestage HD. In nonhuman primates, low doses of 3-NP administered over 3 to 6 weeks cause apomorphine-induced choreaform movements and dystonia of the limbs (Palfi et al. 1996), and when administered over 4 months, spontaneous dyskinesias and dystonia arise (Brouillet et al. 1995). Although motor impairments are a hallmark of HD, it is critical that models also recreate the more debilitating cognitive aspects of the disease, such as impairments in working memory and attention. Toward that end, monkeys treated with 3-NP display cognitive deficits on the object retrieval detour task (ORDT), which purposefully taxes working memory and relies on intact frontostriatal circuitry (Palfi et al. 1996). 3-NP induces striatal toxicity (Figure 2B) by causing the degeneration of GABAergic medium spiny neurons in the striatum, in a pattern similar to the neuronal death seen in HD brains (Beal et al. 1993). Although 3-NP is administered systemically, it causes selective degeneration of neurons in the lateral striatum (Blum et al. 2001), mimicking the severe pathology in the dorsolateral putamen seen in the 361

HD brain (Vonsattel et al. 1985) and suggesting that neurons of the lateral striatum may be more susceptible to mitochondrial damage. 3-NP causes cell death by a combination of necrosis and apoptosis, both of which are seen in the HD brain. Immediately after 3-NP administration there is a wave of necrotic cell death followed by slow apoptosis (Pang and Geddes 1997).

Therapeutic Studies in Toxin Models The profound and predictable cell loss in toxin models makes them ideal for use in neuroprotective and neurorestorative studies. Toxin-induced HD models have tested many therapies, such as coenzyme Q10 and creatine, that are either in use or planned for use in human clinical trials. Ciliary neurotrophic factor (CNTF1), thus far the only trophic factor tested in an HD clinical trial, was evaluated in rodent and toxin models of HD for its ability to prevent neuron loss and alleviate toxin-induced motor and cognitive deficits. Emerich and colleagues (1998) demonstrated that implants of encapsulated cells genetically modified to secrete CNTF obviated the motor and cognitive deficits that resulted from bilateral QA lesions in rats. This group has also demonstrated benefits of gene delivery of trophic factors in primate HD models in which similar grafts prevented striatal cell loss, preserved striatopallidal and striatonigral circuitry, and prevented the atrophy of layer V cortical neurons in QA-lesioned monkeys (Emerich et al. 1997). CNTFexpressing cells transplanted into the 3-NP-lesioned primate striatum improved motor and cognitive function within 3 months after injection (Bachoud-Levi et al. 2000). There was no discernible cell loss in the CNTF-treated lesioned striatum compared to unlesioned control monkeys. Toxinbased models of HD have also been effective for testing other trophic factors (Kordower et al. 2000). Brain-derived neurotrophic factor (BDNF), a compound produced by cortical neurons and secreted in the striatum, is essential for the survival and maintenance of striatal neurons (Canals et al. 2001). BDNF has been tested extensively in rodent toxin models and, more recently, in transgenic models. Cells engineered to produce BDNF completely protect striatal MSNs after grafting into the QA-lesioned rat striatum (Perez-Navarro et al. 2000b). Members of the GDNF family ligands (GFLs), including glial-derived neurotrophic factor (GDNF1) and neurturin, are of great interest in HD animal models due to their important roles in the growth, development, and trophic support of striatal neurons. GDNF and neurturin have differential effects on the two populations of MSNs. GDNF selectively protects the substance P neurons of the direct circuit in a QA rat model of HD (Perez-Navarro et al. 1999), and neurturin selectively protects enkephalin-containing neurons of the indirect circuit (Perez-Navarro et al. 2000a) in this same model. Moreover, studies have shown that delivery of the GDNF or neurturin genes is neuroprotective when administered to 3-NP-lesioned rats. Specifically, ad362

eno-associated viral (AAV1) gene delivery of GDNF or neurturin protects striatal neurons from degeneration and prevents motor deficits on both a raised platform test and a quantitative neurological rating scale used to assess ambulatory capability (McBride et al. 2003; Ramaswamy et al. 2006). Examination of neuroprotective therapies in toxin models may be useful for understanding the benefits of treating patients whose HD diagnosis precedes the onset of cell degeneration or behavioral symptoms. However, surgical therapies such as gene therapy are impractical for presymptomatic clinical trials, given the large number of patients that would be necessary in order to demonstrate a neuroprotective effect. Additionally, most patients do not undergo genetic testing, and consult the clinician only after the onset of clinical symptoms. The development of potent postsymptomatic neurorestorative therapies is therefore critical. Toward this end, human fetal neural stem cells have been shown to attenuate the motor deficits associated with a QA lesion in a rat model of HD (McBride et al. 2004). GDNFexpressing neural stem cells transplanted into a QAlesioned striatum protect neurons from degeneration and alleviate motor deficits (Pineda et al. 2007). In a different QA model, rats that received transplants of fetal cortical cells administered intravenously through the tail vein showed improved rotational behavior and had decreased striatal atrophy in response to migration of cells to the lesioned striatum (Lee et al. 2005). HD is a promising arena for the development of stem cell therapy. In light of the links between HD and mitochondrial insufficiency (discussed above), the 3-NP model can efficiently test therapies that restore mitochondrial function. Coenzyme Q10 is a molecule in the electron transport chain that carries electrons from complexes I and II to complex III (Kidd 2005) and that reduces oxidative stress by keeping electrons in the mitochondrial membrane. Coenzyme Q10 can reverse the energy deficits associated with 3-NP lesions (Kasparova et al. 2006). Creatine, a compound capable of buffering ATP levels in the cell, is the focus of clinical trials for HD based on promising motor and cognitive benefits obtained in the 3-NP rat model (Shear et al. 2000).

Pitfalls of Toxin Models Although both excitotoxins and mitochondrial inhibitors have been reliable models for specific aspects of HD, there are a number of drawbacks to their use. First, there is no clear association between their mechanism of action and the genetic cause of HD. As such, there is no production and misfolding of mutant htt, along with the subsequent development of neuronal and cytoplasmic inclusions. There is also intense debate about whether the formation of cellular inclusions is a protective mechanism to sequester a toxic protein or a more direct cause of cellular dysfunction. However, it remains clear that their formation is an integral pathological feature of the disease and, as such, should be ILAR Journal

replicated in a faithful animal model. Furthermore, cell death in HD is progressive and its age of onset inversely proportional to the number of CAG repeats in the mutant htt protein (Andrew et al. 1993), whereas cell death in the toxin models is immediate and not progressive. Additionally, while the neuronal death that occurs in HD likely involves excitotoxic and metabolic mechanisms similar to those seen in toxin-based models, these degenerative processes are likely secondary to factors such as the accumulation of mutant htt. Furthermore, symptoms in HD are progressive and unfold over the course of several years, and cognitive symptoms often arise before the onset of motor and personality disturbances and are more disruptive (Hahn-Barma et al. 1998). Although the models described above replicate some of the cognitive changes, it is not possible to evaluate personality changes such as depression, suicidal tendencies, mania, and obsessive-compulsive behavior (Huntington 2003) in animals. Finally, motor symptoms in HD progress from hyperkinetic in early disease to hypokinetic in later stages, and careful adjustments in the dose of 3-NP cause animals to manifest either hyperkinetic or hypokinetic movements. But while the massive cell death produced by QA and 3-NP makes these good models to study neuroprotective and neurorestorative therapies for HD, it makes them inefficient to study the progressive nature of the disease.

Genetic Models With the discovery of the htt mutation in 1993, it became possible to create animal models with a similar genetic background of the disease seen in humans with HD (The Huntington’s Disease Collaborative Research Group 1993). Although genetic models of HD have used lower invertebrates such as fruit flies and nematodes, genetic mice created to express the mutant htt gene, or a portion thereof, have been the most commonly used models to answer basic biological questions about the disease and to investigate potential therapies. There are two main categories of genetic mouse models, transgenic and knock-in. Transgenic mice result from the random insertion of a portion of the human htt gene, containing the polyglutamine repeat, in the mouse genome, the expression of which can be driven by different promoters (Table 2). Alternatively, “knocking in” a portion of the human htt gene in the mouse htt gene locus on chromosome 7 results in the creation of knock-in mice (Table 3). The exogenous mouse htt promoter drives expression of the mutant htt protein in these mice and so production of the mutant protein is both spatially and temporally accurate. There are several transgenic and knock-in rodent models, and we discuss them below.

Transgenic Rodent Models R6/2 The R6/2 is the most commonly used transgenic mouse model of HD. Bates and colleagues, in their landmark study, Volume 48, Number 4

2007

created this transgenic murine line by inserting a 1.9-kB fragment derived from the 5⬘ end of the human htt gene into the mouse genome (Mangiarini et al. 1996). This fragment contains only exon 1 of the human htt gene and expresses approximately 144 CAG repeats. The truncated mutant htt gene is randomly inserted into the mouse genome and the mouse expresses three copies of the htt gene—two of its own wild-type copies and one mutant human copy. Expression of the mutant htt gene is driven by the human htt promoter. The mutant gene is expressed in all cells of the mouse and at 75% of the level of the wild-type gene (Mangiarini et al. 1996). The large number of repeats in the R6/2 model corresponds to a juvenile onset of HD symptoms in humans. This model has a very aggressive behavioral phenotype. Some R6/2 mice have been known to develop symptoms as early as 4 weeks, although the average age at onset of symptoms is 9 to 11 weeks. Average age at death is 10 to 13 weeks, and animals rarely survive past 14 weeks. Initial studies failed to demonstrate significant pathology in the R6/2 mice (Reddy et al. 1998; Sun et al. 2002), with the exception of numerous htt-containing inclusions in the striatum and cortex. However, a recent study by Stack and colleagues (2005) investigated the time-dependent changes in R6/2 mice with respect to brain volume, striatal volume, and striatal neuronal counts. This investigation revealed a time-dependent reduction in brain weight and brain volume beginning at postnatal days 30 and 60, respectively. Critically, significant striatal neuronal atrophy and loss occurred at 90 days of age. Pathological changes were associated with changes in body weight, rotarod performance, grip strength, and dystonia. Furthermore, Sun and colleagues (2002) compared cells expressing enkephalin and substance P in a localized area of the striatum. There was a significant reduction in pre-proenkephalin (an enkephalin marker) messenger RNA (mRNA1) in the striatum starting at week 6, but no change in pre-protachykinin (a substance P marker) mRNA at any age. There was also a significant reduction (>50%) in the number of enkephalin-expressing neurons in the striatum at 12 weeks of age but no change in the number of substance P-expressing neurons. Mice develop progressive mutant htt-positive inclusions in the striatum and, by 8 weeks of age, the majority of neurons in the brain contain inclusions. Inclusions initially appear in the cortex and the hippocampus (CA1 before CA3) and then progress to the striatum. The exact mechanism of cell death in the R6/2 brain is unknown, but one study has reported the presence in the R6/2 brain of “dark neurons” resembling those in the human HD brain (Turmaine et al. 2000). This type of cell death appears to be neither apoptotic nor necrotic and occurs in the striatum, cingulate cortex, and cerebellum. Although the number of CAG repeats displayed by R6/2 mice (144) models that of juvenile HD cases, the structural changes in the striatum mimic adult-onset HD. In this regard, mRNA for enkephalin is significantly reduced in the striatum compared to mRNA for substance P. Additionally, enkephalin projections to the globus pallidus exhibit degeneration while substance P projections to the 363

364

ILAR Journal

Transgenic mouse

Transgenic mouse

Transgenic mouse

Transgenic mouse

Rat

R6/1

N171-82Q

YAC

Transgenic rat

Species

R6/2

Animal model

1962 base pairs from the N-terminal of the rat htt gene (von Horsten et al. 2003)

First 171 a.a. YAC of human htt randomly inserted into mouse genome (Schilling et al. 1999) Yeast artificial chromosome expressing entire human htt protein (Hodgson et al. 1999)

First 90 a.a. of human htt randomly inserted into mouse genome (Mangiarini et al. 1996)

First 90 amino acids (a.a.) of human htt randomly inserted into mouse genome (Mangiarini et al. 1996)

Construct

Table 2 Transgenic models

Endogenous rat htt promoter

Human htt promoter

Prion

Human huntingtin

Human huntingtin

Promoter

51 humanderived repeats

72, 128

82

116

144

CAG repeat size 90 d: Significant neuronal loss and neuronal atrophy in striatum (Stack et al. 2005). 12 wk: Significant loss of enkephalin positive neurons in the striatum (Sun et al. 2002). Inclusions throughout the brain 5 M: Decreased DARPP-32 in the striatum beginning at 5 months of age. Decrease in striatal volume and presence of cellular inclusions (van Dellen et al. 2000) 16 wk: 25% neuronal loss in striatum. 20% cell shrinkage. Inclusions in striatum, hippocampus, and cortex (McBride et al. 2006) 12 M: 18%–40% cell loss with majority of loss occurring in the lateral striatum. Increased nuclear htt staining. 18 M: Inclusions in the YAC 128 strain (Hodgson et al. 1999; Slow et al. 2003) 8 M: Enlarged lateral ventricles. 12 M: Inclusions throughout the brain

Cells affected

1-2 M: Transgenic rats perform better on the rotarod test compared to wild-type littermates but progressively decline thereafter. 10-15 M: Gait abnomalities and dyskinesias of the head. 12 M: Slower in traversing an elevated beam. 24 M: 20% weight loss compared to wild-type rats (Nguyen et al. 2006)

3 M: Hyperkinesia on an open-field test. 6 M: hypokinesia on open-field test. 12 M: 50% decrease in body weight compared to the wild-type littermates. Pronounced circling behavior, gait abnormalities, ataxia, and hindlimb clasping. Progressive decline on the rotarod test.

11 wk: Rotarod deficits, clasping behavior, weight loss (McBride et al. 2006)

22 wk: Body weights plateau after which they begin to decline. Gait abnormalities as measured by footprint analysis and hindlimb clasping behavior (Naver et al. 2003)

40 d: Decline on the rotarod test. 9 wk: resting tremor, chorea-like movements, stereotypies, clasping behavior, epileptic seizures, narcolepsy, weight loss, and spontaneous shuddering movements (Mangiarini et al. 1996; Stack et al. 2005; Luesse et al. 2001)

Motor symptoms

12 M: Deficits on radial arm maze and elevated plus maze

Deficits on T-maze (Van Raamsdonk et al. 2005)

14 wk: Deficits on radial arm water maze text of reference and working memory (Ramaswamy et al. 2004)

Decreased anxiety on open-field test

20 d: Increased exploratory behavior in open-field test. 60 d: Decreased exploratory behavior on open-field test. 3.5 wk: Deficits on Morris water maze and T-maze (Lione et al. 1999)

Cognitive symptoms

Table 3 Knock-in models

Promoter

CAG repeat size Cells affected

Species

Construct

HdhQ92 mouse (Wheeler et al. 1999, 2000)

Replacing exon 1 of the mouse huntingtin htt gene with a mutant human exon 1

mouse htt promoter

92

HdhQ111 mouse (Wheeler et al. 1999, 2000)

Replacing exon 1 of the mouse huntingtin gene with a mutant human exon 1

mouse htt promoter

111

CAG140 mouse (Menalled et al. 2003)

Inserting CAG repeats into the mouse htt gene

mouse htt promoter

140

CAG15O mouse (Lin et al. 2001; Yu et al. 2003)

Inserting CAG repeats into the mouse htt gene

mouse htt promoter

150

pallidus and substantia nigra are relatively preserved. Before the onset of symptoms, there is a loss of different neurotransmitter receptors similar to that seen in adult HD, such as the type 1 metabotropic glutamate receptor, D1 and D2 dopamine receptors, and muscarinic cholinergic receptors (Cha et al. 1998). R6/2 mice show multidomain deficits in motor and cognitive function. Motor symptoms include resting tremor, chorea-like movements, stereotypic involuntary grooming movements, and dystonia of the limbs when suspended by the tail (clasping behavior) (Mangiarini et al. 1996; Stack et al. 2005). The mice show progressive decline on the Volume 48, Number 4

2007

No striatal degeneration. 4.5 M: Translocation of mutant huntingtin protein to the nucleus No striatal degeneration. 4.5 M: Huntingtin protein translocates to the nucleus and appears punctuate. 24 M: Striatal gliosis 2 M: Nuclear and neuropil inclusions in the striatum, cortex, hippocampus, and cerebellum

14 M: Significant increase in striatal gliosis compared to wild-type littermates. The striatum shows an increase in EM48 positive nuclear aggregates. Breakdown of myelin

Motor symptoms

Cognitive symptoms

No overt symptoms

No overt symptoms

24 M: Gait abnormalities

No overt symptoms

1 M: Increase in rearing behavior using both forelimbs and open-field test locomotion compared to wild-type mice which decrease at 4 months. 12 M: Decrease in stride length 4 M: Onset of progressive deficits on the rotarod, clasping phenotype, hypoactivity, and gait disturbances. 25 M: Significantly smaller than their wild-type littermates

No overt symptoms

No overt symptoms

rotarod test starting as early as 40 days of age and by 12 weeks are unable to maintain their balance for even 10 seconds (Luesse et al. 2001). In an open-field test of locomotion and exploratory behavior, R6/2 mice show increased exploratory behavior compared with wild-type mice at 20 days. This hyperactivity progressively declines and by 60 days of age they are hypoactive relative to wildtype mice. R6/2 mice also suffer from epileptic seizures and spontaneous shuddering movements. Their body weight plateaus at week 9 and at the end of their life the mice weigh approximately 70% less than their wild-type littermates (Mangiarini et al. 1996). Like patients with HD, R6/2 365

mice show a dramatic loss of orexin-positive neurons in the hypothalamus associated with narcolepsy (Petersen et al. 2005), although the function of this degenerative change is unclear. The R6/2 mice also show cognitive deficits (Lione et al. 1999). As early as 3.5 weeks, before the onset of overt motor symptoms, they reveal deficits in the Morris water maze spatial learning task, and older animals (7 to 8 weeks) completely lose the ability to learn this task, partly due to the deterioration of swimming ability. Animals also perform poorly on the alternating T-maze task and cannot learn to alternate entry into opposite arms. This may occur because these mice, similar to HD patients, have a tendency to perseverate in relation to the original stimulus. Although these cognitive deficits are evident relative to wild-type controls, it is unclear whether they are related to those seen in HD. Deficits on the Morris water maze, which are typically spatial in nature and hippocampal dependent, may be due to accumulation of htt aggregates in the hippocampus. In contrast, cognitive deficits seen in HD are typically frontostriatal and are characterized by losses in executive function, procedural memory, and psychomotor skills (Bylsma et al. 1990; Heindel et al. 1988; Lange et al. 1995). Deficits in the reversal phase of a T-maze are related to those seen in HD patients who participate in the Wisconsin Card Sorting Task. Both tests require subjects to replace a previously learned strategy using their intact frontostriatal circuitry (Goldberg et al. 1990; Van Raamsdonk et al. 2005). Therefore deficits in the T-maze may be more closely related to deficits seen in HD. R6/1 The method of creation for R6/1 mice is similar to that for R6/2 mice except that they contain only 116 repeats, making their behavioral phenotype relatively mild (Mangiarini et al. 1996). As with the R6/2 mice, mutant huntingtin gene expression is driven by the human htt promoter in all cells of the body, but it is at only 31% of the expression levels seen in the endogenous mouse gene (compared to 75% in the R6/2 line). There is no loss of neuronal nuclei (NeuN) stained or met-enkephalin immunopositive cells in the striatum at any age (Naver et al. 2003). However, R6/1 mice do show decreased dopamine and cyclic adenosine monophosphate (cAMP)-regulated phosphoprotein (DARPP-32) staining in the striatum beginning at 5 months of age (van Dellen et al. 2000), indicating cellular dysfunction. There is also evidence of striatal shrinkage in the absence of cell death. Cellular inclusions appear throughout the brain and their numbers increase with age as in the R6/2 model. Body weights of these mice plateau at 22 weeks, after which they begin to decline (Naver et al. 2003). The animals also show other deficits like gait abnormalities as measured by footprint analysis, decreased anxiety, and hindlimb clasping behavior. The R6/1 model is not used as often as the R6/2 model and is therefore less well understood. 366

N171-82Q Borchelt and colleagues created the N171-82Q transgenic mouse model by inserting the first 171 amino acids from the N-terminal of the human htt gene into the mouse genome (Schilling et al. 1999). Expression of the N-terminal fragment of htt in this model is driven by the mouse prion promoter and, as such, mutant htt is expressed throughout the mouse brain. But its expression is restricted to neurons and is not found in glia. As with the R6 mice, these transgenic mice contain two normal (wild-type) copies of the htt genes along with one mutated copy of the truncated human htt gene. This model has fewer (82) polyglutamine repeats relative to the R6 mice, resulting in a later onset of symptoms, which may make it more relevant for modeling adultonset HD. We note, however, that line 81 of this mouse model (commercially available through Jackson Laboratories) has a shortened life span (5 to 6 months). For the first 2.5 months of life, these animals appear completely normal in their behavior. Once the behavioral syndrome begins, animals display resting tremor, hypokinesia, hindlimb and forelimb clasping, and abnormal gait. Beginning at week 11 they also show a progressive decline in performance on the rotarod test as well as the onset of clasping (McBride et al. 2006). With respect to cognition, the N171-82Q mice show deficits in the radial arm water maze test of working and reference memory at 14 weeks (Ramaswamy et al. 2004). However, as in previously described studies, this cognitive deficit may be due to the numerous htt inclusions in the hippocampus rather than frontostriatal circuitry. In contrast to the R6/2 model, these mice do not display hyperkinetic movements. Investigators have reported neuronal cell loss in the striatum of N171-82 HD mice at 16 weeks of age—a 25% neuronal loss in the striatum and a 20% decrease in striatal cell volume (McBride et al. 2006). N171-82Q mice also exhibit mutant htt-positive inclusions in the striatum, cortex, and hippocampus at 16 to 20 weeks. The later onset of behavioral symptoms and striatal neurodegeneration and the more protracted time course of symptoms relative to the R6/2 mice make this an attractive model for the study of presymptomatic therapies. These features also allow for a longer experimental window during which therapies can be administered before the pathological sequelae of the disease commence. YAC Mice YAC transgenic mice are the creation of Hayden and colleagues, who used a yeast artificial chromosome (YAC) vector system to express the entire human htt gene under control of the human htt promoter (Hodgson et al. 1999). YAC mouse strains contain either 72 or 128 CAG repeats. Both strains show a decrease in the number of neurons, preferentially in the lateral striatum (Hodgson et al. 1999; Slow et al. 2003). YAC 72 mice show a 50% decrease in body weight compared to the wild-type littermates at 12 ILAR Journal

months. They also demonstrate pronounced circling behavior, gait abnormalities, ataxia, and hindlimb clasping. YAC 128 mice show a progressive decline on the rotarod test and are hyperkinetic on an open-field test beginning at 3 months, but then begin to show signs of hypokinesia at 6 months. Both the YAC 72 and 128 models have increased nuclear htt staining. Only the YAC 128 striatum shows positive staining for inclusion bodies at 18 months, whereas the YAC 72 model very rarely shows the presence of huntingtin aggregates. Studies of cognitive dysfunction in the YAC 128 model have used a simple linear swimming chamber and a swimming T-maze test at 8.5 months of age (Van Raamsdonk et al. 2005). The simple linear swimming test involves placing the mouse in the center of a linear chamber facing away from an escape platform that is located at one end of the chamber. Evaluation of procedural learning depends on the mouse’s ability to turn immediately and swim toward the platform. YAC 128 mice show severe deficits in the performance of this task at 8 months of age: they fail to turn immediately and they take significantly longer than wild-type mice to locate and reach the platform. In a second cognitive test using a swimming T-maze, the mice first learn to swim to a platform in the right arm of a T-shaped chamber, then the platform is moved to the left arm and the animals have to locate it. Wild-type mice initially reenter the right arm looking for the platform and, when they fail to locate it, immediately swim to the left arm. YAC 128 mice swim toward the right arm but either return to the start arm or take significantly more trials to learn that the platform has been relocated to the left arm. Their failure in this task indicates a deficit similar to that seen in humans: they perseverate based on what they learned earlier instead of forming a new strategy. Because YAC mice live much longer than their R6 and N171-82Q transgenic counterparts, the YAC mouse model is an attractive candidate for longterm therapeutic studies. Transgenic Rats The only transgenic rat model currently in use is the creation of von Horsten and colleagues (2003). These rats express 51 human-derived CAG repeats inserted in a 1962 bp rat htt fragment of complementary DNA (cDNA). Expression of this gene is under control of an 885 bp endogenous rat promoter. The relatively smaller number of CAG repeats results in an adult-onset progressive phenotype. The mutant htt fragment is expressed mainly in the basal ganglia, frontal and temporal cortices, hippocampus, and midbrain, and at much lower levels in the cerebellum and spinal cord. Like other transgenic models these rats appear normal at birth but by 24 months of age weigh 20% less than wild-type rats. Motor symptoms appear at 10 to 15 months of age but are limited to gait abnormalities and dyskinesias of the head. Interestingly, transgenic rats perform better on the rotarod test for the first 2 months of age compared to wild-type littermates but progressively decline thereafter (Nguyen et Volume 48, Number 4

2007

al. 2006). Transgenic rats are also slower in traversing an elevated beam at 12 months. Cognitive deficits arise around the same age on the radial arm maze and elevated plus maze. Inclusions occur diffusely throughout the brain and start to develop at 12 months of age. Magnetic resonance images reveal enlarged lateral ventricles at 8 months, indicating striatal atrophy. The benefit of using rats as models is that they have a longer life span compared to mice, allowing for their use in long-term therapeutic interventions.

Pitfalls of Transgenic Models Some researchers believe that because transgenic mice contain multiple copies of the htt gene (two normal mouse copies and one mutated human copy) they are not representative of the genetics of the disease as it appears in humans, who usually have only one normal copy of the htt gene and one mutated copy. Also, the presence of either the fulllength or truncated form of the human htt gene in the mouse genome may cause additional pathology not associated with the disease. In transgenic models, the random insertion of the human htt gene in the mouse genome may interfere with the normal function of other genes not related to HD. Moreover, the expression of htt is driven by an artificial promoter, whose spatial and temporal control of expression is different from that of the endogenous mouse htt promoter. Additionally, depending on the specific transgenic model and the promoter that drives expression, htt production in transgenic mice does not necessarily mirror production in the same cell types that appear in human HD. For example, in the N171-82Q model, htt is driven from a prion promoter, restricting expression solely to neurons and not glia. But in human HD, htt is expressed in both neurons and glia, and recent evidence from an in vitro study demonstrates that htt expression in glia leads to the formation of astrocytic inclusions and the inability to sequester synaptic glutamate, a major contributing factor to neuronal excitotoxicity (Shin et al. 2005). These concerns spurred the creation of knock-in mouse models (Table 3).

Knock-in Mouse Models Knock-in mouse models are considered the most accurate for HD from a genetic standpoint. Creation of these mice involves replacing a portion of the mouse htt gene with a mutant human copy that contains an expanded CAG region. Thus these mice have only two copies of the htt gene—a wild-type allele and a mutant allele, both under control of the mouse htt promoter. Initial studies in knock-in mice were disappointing because the mice did not show the overt neuropathology or symptoms seen in the transgenic mice. But closer examinations have revealed subtle changes in mice with larger numbers of repeats. The chimeric HdhQ lines were the first to be developed by replacing exon 1 of the mouse htt gene with a mutated exon 1 containing either 367

111 or 92 CAG repeats in the HdhQ111 and HdhQ92 lines, respectively. Lines with large polyglutamine repeats cause a CAG repeat instability. Thus, similar to what occurs in humans, the repeat length increases in subsequent mouse generations, predisposing them to an earlier, potentially juvenile onset of symptoms (Wheeler et al. 1999). Although there is no overt striatal degeneration in either of these lines, striatal pathology is evident. Htt protein translocates to the nucleus at age 4.5 months in both HdhQ111 and HdhQ92 lines and appears punctuate in the HdhQ111 mouse (Wheeler et al. 2000). Additionally, there is an increase in gliosis in the striatum at 24 months of age. Despite moderate pathology in the brain, the HdhQ111 mice do not exhibit any overt behavioral pathology. Evaluations of breeding, body weight, brain weight, life span, hindlimb clasping, exploratory behavior, or rotarod performance reveal no significant difference between the knock-in models and wildtype littermates. Scott Zeitlin’s group created the CAG140 knock-in mouse, which expresses 140 CAG repeats (Menalled et al. 2003). CAG140 mice show minor hypoactivity at 1 month of age, as seen using an open-field test, but at 12 months motor dysfunction is more apparent, as the mice show overt gait abnormalities measured by a decrease in stride length. CAG140 knock-in mice also show a significant increase in rearing behavior using both forelimbs compared to wildtype mice at 1 month of age, but this begins to decline significantly at 4 months. Although the mice do not show significant rotarod deficits up to 5 months of age, such deficits become evident at 9 months compared to wild-type littermate controls (McBride and Davidson, unpublished observations). Inclusion bodies appear in CAG140 knock-in mouse brains—in the striatum, cortex, hippocampus, and cerebellum—starting at 2 months, and progressively increase in both size and number. These aggregates form in both the nucleus and neuropil. Neuronal cell loss has not yet been quantified in these mice. Perhaps the most promising of the knock-in mice is the CAG150 mouse model, created by Detloff and colleagues (Lin et al. 2001). These mice show progressive deficits on the rotarod, a clasping phenotype, hypoactivity, and gait disturbances starting at 4 months. They are also significantly smaller than their wild-type littermates after 25 weeks of age. At 14 months, they have a significant increase in striatal gliosis compared to their wild-type littermates (Yu et al. 2003), and the striatum shows an increase in nuclear aggregates. Although there was a complete absence of apoptotic TUNEL staining in the brain, electron microscopy revealed dark bodies surrounding cytoplasmic vacuoles, evidence of cellular dysfunction. Electron microscopy also revealed degenerating axons as indicated by the presence of dark, swollen organelles and by myelin disrupted with htt aggregates. The more severe pathology seen with the CAG150 knock-ins relative to the trivial pathology in CAG140 knock-ins suggests that future models using even higher CAG repeats may be able to induce pathology more representative of the human disease. 368

Pitfalls of Knock-in Mice Knock-in mice are, in some respects, the most representative models of htt protein pathology. However, their behavioral deficits are not as pronounced as those seen in transgenic mice and, when they are present, they usually take much longer to develop. Few reports have characterized the brain pathology in these models, and most of these reports described only the presence of cellular inclusions and striatal astrogliosis without demonstrating frank neuronal loss, although there are some descriptions of cellular dysfunction and degenerating axons in the CAG150 knockin model. There is also no report of reduced life span in any knock-in mouse strain. The lack of substantive neuropathology makes it difficult to study the effects of therapeutic interventions in this model.

Gene Delivery of Mutant Huntingtin A nontransgenic approach to creating a genetic model of HD has recently been established. This model involves the site-specific induction of overexpression of mutant htt using viral vectors. The approach bypasses potential developmental plasticity that can occur in transgenic and knock-in models in terms of genetic drift, which leads to larger numbers of repeats than desired in successive generations of mice. Déglon and colleagues first examined the effects of lentiviral delivery of a truncated form of the human htt protein with an expanded polyglutamine region (82 repeats) in rats (de Almeida et al. 2002). As early as 1 week after lentiviral transfection, cells in the striatum begin to express EM48 positive inclusions, and the number of inclusions increased progressively during the 4 weeks after injection. Neuronal degeneration was evident in the injected striatum, accompanied by a decrease in DARPP-32 immunostaining. The neurodegeneration was specific to the GABAergic neurons, while interneurons were not affected. Because the creation of transgenic nonhuman primates is both time- and costinefficient, this model is particularly attractive for examining complex cognitive and motor behaviors that can best be studied in nonhuman primates. Preliminary data from Palfi and colleagues (personal communication) indicate that such an approach can be successful for establishing motor deficits analogous to HD, and additional studies seek to determine the effects of lentiviral expression of the mutant htt protein on cognitive symptoms in nonhuman primates.

Therapeutic Strategies in Genetic Models Genetic models may not only help researchers establish the basic biological principles of HD pathology but also enable the study of effects of potential therapies in models that more closely mirror the human disease pheno- and genotype. For example, coenzyme Q10 is in clinical trials after success in the R6/2 transgenic mouse model (Smith et al. ILAR Journal

2006). A dose-response study in R6/2 mice showed that high doses of coenzyme Q10 (1,000-20,000 mg/kg/day) prevented weight loss, improved motor deficits on the rotarod and grip strength tests, and prolonged survival (Smith et al. 2006). Tests of coenzyme Q10 potentiate its beneficial effects in the R6/2 model in combination with the antiinflammatory agent minocycline (Stack et al. 2006) and the antioxidant vitamin E (Kasparova et al. 2006). A study of the effects of coenzyme Q10 and remacemide (an NMDA receptor antagonist) in the N171-82Q transgenic model reported significant, but transient, improvements in performance on the rotarod test and attenuation of weight loss (Schilling et al. 2001). There was, however, no increase in life span or alteration in neurodegenerative processes in the brain. Studies have also tested the effects of creatine in the R6/2 transgenic mouse. By reversing the reduced levels of ATP in the brain, creatine slowed the progression of motor symptoms and increased survival in this model when administered early and mid-course in the disease (Dedeoglu et al. 2003), and it diminished weight loss and decreased brain atrophy. A recent assessment of the neuroprotective effects of trophic factors in genetic mouse models indicates that AAV gene delivery of GDNF is beneficial in reversing motor deficits and neuropathological sequelae in the N171-82Q transgenic mouse model (McBride et al. 2006). Intrastriatal injections of AAV-GDNF in presymptomatic N171-82Q mice delayed the onset of motor symptoms observed on rotarod and clasping tests and prevented neurodegeneration and striatal atrophy compared to control-injected transgenic littermates (McBride et al. 2006). Preliminary results suggest similar findings after the administration of AAVneurturin, GDNF’s functional analogue (Ramaswamy et al. 2006). Interestingly, lentiviral delivery of GDNF in the R6/2 transgenic model did not prevent the onset of behavioral symptoms or the development of neuronal inclusions in the striatum (Popovic et al. 2005). The more aggressive behavioral phenotype seen in the R6/2 model, coupled with delivery of the AAV-GDNF in the presymptomatic phase of the syndrome in N171-82Q mice, may explain the differences in these two studies. RNA interference (RNAi) is a recently developed and exciting therapy for the treatment of HD. It uses short interfering RNA (siRNA), short hairpin RNA (shRNA), or microRNA (miRNA) sequences to bind to and reduce the expression of complementary messenger RNA (mRNA) sequences. Specific to HD, use of this therapy can decrease the expression of mutant htt and the subsequent sequelae of pathological events that result from its production. Using this procedure in the N171-82Q transgenic mouse model, viral delivery of shRNA directed against a sequence in exon 2 of the mutant htt gene showed a 50-55% decrease in the mutant htt protein in the injected striatum and prevented the formation of mutant htt-positive inclusions in transduced cells (Harper et al. 2005). There was also a rescue of motor deficits on the accelerating rotarod test as well as the gait test. Another study demonstrated that ventricular delivery of Volume 48, Number 4

2007

siRNA directed against htt reduced both brain atrophy and the presence of htt-positive neuronal inclusions in the R6/2 transgenic mouse model (Wang et al. 2005). This study also reported a rescue of motor deficits (rotarod, clasping, and open-field tests) and an increase in survival time. The above-mentioned studies all examined the effects of RNAi in presymptomatic models, meaning that the therapy was administered before the onset of functional and anatomical decline. A recent encouraging study reported that shRNA administration in postsymptomatic HD190QG transgenic mice (htt-EGFP fusion mice) reduced the number of httpositive inclusions in the striatum, suggesting that this therapy can be effective in not only preventing but also reversing certain aspects of HD pathology (Machida et al. 2006). This study also suggests that inclusion bodies are perhaps more dynamic than previously speculated, and that if the burden of mutant htt production is lessened cells may be able to effectively clear already formed inclusion bodies, an exciting prospect for HD therapy. Yamamoto and colleagues (2000) reported a similar finding, using a conditional model that controlled levels of htt expression by a tetracycline-regulated system. Although early studies demonstrated that htt knock-outs are lethal at the embryonic stage (Nasir et al. 1995), it remains unknown whether a partial reduction of htt in the adult knock-in mouse striatum will elicit beneficial effects.

Considerations for Animal Use HD is a severely debilitating disease. Animal models can mimic symptoms seen in humans either by using specific neurotoxins such as QA or 3-NP or through genetic engineering. The symptoms most commonly associated with the diseased state induced by the neurotoxins are loss of weight, possible tremor or seizures, eventual paralysis, recumbence, and often death. Similar symptoms, including weight loss, hypokinesia, recumbence, and premature death, appear in many genetic models in later stages of the disease. These symptoms closely mimic those seen during the progression of the diseased state in humans and are therefore appropriate for inclusion in animal models to test specific hypotheses. Proper care of animals in such models should ensure close daily monitoring and weekly weight checks. If pain is observed, the animals should receive analgesics as appropriate, to prevent potential discomfort related to manifestation of the diseased state. The protocol should include consultations with veterinary staff when necessary as well as appropriate treatment. Some animals may require gavage if they become unable to feed themselves. Immediately after toxin administration or in later stages in genetic models, euthanasia may be necessary for moribund animals.

Conclusions Rodent models are the most widely used to study HD pathogenesis and potential therapies. Nongenetic models induce 369

cell death either by excitotoxicity or by disruption of mitochondrial machinery. Perhaps the most important development in HD research since the discovery of the gene itself is the creation of a variety of genetic models. Genetic rodent models of HD are created by introducing genes that express the mutated htt protein (transgenics) or by replacing the mouse gene or a portion of it with a mutated gene (knockins). These are the “gold-standard models” of HD, but due to the presence of overt frontostriatal cognitive deficits and the lack of substantial striatal degeneration, no current model can claim to be the best. Like all animal model experiments, the choice of a particular model depends on the question under investigation. Toxin-based models may still have a role, but most experimental hypotheses, especially those involving therapeutic interventions, require success in a genetic model, the choice of which again depends on the experimental question. What is absolutely clear is that additional model development is still needed. New knock-in models may better recapitulate the pathological and functional deficits seen in clinical HD. Finally, vector-delivered htt has an important role in HD research, as it enables the study not only of a genetically based pathological process in nonhuman primates but also of the specific features of HD that are the most debilitating in species more closely linked to humans.

References Albin RL, Reiner A, Anderson KD, Penney JB, Young AB. 1990. Striatal and nigral neuron subpopulations in rigid Huntington’s disease: Implications for the functional anatomy of chorea and rigidity-akinesia. Ann Neurol 27:357-365. Albin RL, Young AB, Penney JB. 1989. The functional anatomy of basal ganglia disorders. Trends Neurosci 12:366-375. Alexander GE, Crutcher MD. 1990. Functional architecture of basal ganglia circuits: Neural substrates of parallel processing. Trends Neurosci 13:266-271. Alston TA, Mela L, Bright HJ. 1977. 3-Nitropropionate, the toxic substance of Indigofera, is a suicide inactivator of succinate dehydrogenase. Proc Natl Acad Sci U S A 74:3767-3771. Andrew SE, Goldberg YP, Kremer B, Telenius H, Theilmann J, Adam S, Starr E, Squitieri F, Lin B, Kalchman MA. 1993. The relationship between trinucleotide (CAG) repeat length and clinical features of Huntington’s disease. Nat Genet 4:398-403. Bachoud-Levi AC, Déglon N, Nguyen JP, Bloch J, Bourdet C, Winkel L, Remy P, Goddard M, Lefaucheur JP, Brugieres P, Baudic S, Cesaro P, Peschanski M, Aebischer P. 2000. Neuroprotective gene therapy for Huntington’s disease using a polymer encapsulated BHK cell line engineered to secrete human CNTF. Hum Gene Ther 11:1723-1729. Beal MF. 1992. Does impairment of energy metabolism result in excitotoxic neuronal death in neurodegenerative illnesses? Ann Neurol 31: 119-130. Beal MF, Brouillet E, Jenkins BG, Ferrante RJ, Kowall NW, Miller JM, Storey E, Srivastava R, Rosen BR, Hyman BT. 1993. Neurochemical and histologic characterization of striatal excitotoxic lesions produced by the mitochondrial toxin 3-nitropropionic acid. J Neurosci 13:41814192. Beal MF, Ferrante RJ, Swartz KJ, Kowall NW. 1991. Chronic quinolinic acid lesions in rats closely resemble Huntington’s disease. J Neurosci 11:1649-1659. Beal MF, Kowall NW, Ellison DW, Mazurek MF, Swartz KJ, Martin JB.

370

1986. Replication of the neurochemical characteristics of Huntington’s disease by quinolinic acid. Nature 321:168-171. Beal MF, Mazurek MF, Ellison DW, Swartz KJ, McGarvey U, Bird ED, Martin JB. 1988. Somatostatin and neuropeptide Y concentrations in pathologically graded cases of Huntington’s disease. Ann Neurol 23: 562-569. Blum D, Gall D, Cuvelier L, Schiffmann SN. 2001. Topological analysis of striatal lesions induced by 3-nitropropionic acid in the Lewis rat. Neuroreport 12:1769-1772. Bordelon YM, Chesselet MF, Nelson D, Welsh F, Erecinska M. 1997. Energetic dysfunction in quinolinic acid-lesioned rat striatum. J Neurochem 69:1629-1639. Borlongan CV, Koutouzis TK, Freeman TB, Hauser RA, Cahill DW, Sanberg PR. 1997. Hyperactivity and hypoactivity in a rat model of Huntington’s disease: The systemic 3-nitropropionic acid model. Brain Res Brain Res Protoc 1:253-257. Borlongan CV, Randall TS, Cahill DW, Sanberg PR. 1995. Asymmetrical motor behavior in rats with unilateral striatal excitotoxic lesions as revealed by the elevated body swing test. Brain Res 676:231-234. Borlongan CV, Saporta S, Sanberg PR. 1998. Intrastriatal transplantation of rat adrenal chromaffin cells seeded on microcarrier beads promotes long-term functional recovery in hemiparkinsonian rats. Exp Neurol 151:203-214. Brignull HR, Morley JF, Garcia SM, Morimoto RI. 2006. Modeling polyglutamine pathogenesis in C. elegans. Methods Enzymol 412:256-282. Brouillet E, Hantraye P, Ferrante RJ, Dolan R, Leroy-Willig A, Kowall NW, Beal MF. 1995. Chronic mitochondrial energy impairment produces selective striatal degeneration and abnormal choreiform movements in primates. Proc Natl Acad Sci U S A 92:7105-7109. Browne SE, Bowling AC, MacGarvey U, Baik MJ, Berger SC, Muqit MM, Bird ED, Beal MF. 1997. Oxidative damage and metabolic dysfunction in Huntington’s disease: Selective vulnerability of the basal ganglia. Ann Neurol 41:646-653. Burns LH, Pakzaban P, Deacon TW, Brownell AL, Tatter SB, Jenkins BG, Isacson O. 1995. Selective putaminal excitotoxic lesions in nonhuman primates model the movement disorder of Huntington disease. Neuroscience 64:1007-1017. Butters N, Salmon D, Heindel WC. 1994. Specificity of the memory deficits associated with basal ganglia dysfunction. Rev Neurol (Paris) 150: 580-587. Butters N, Wolfe J, Martone M, Granholm E, Cermak LS. 1985. Memory disorders associated with Huntington’s disease: Verbal recall, verbal recognition and procedural memory. Neuropsychologia 23:729-743. Bylsma FW, Brandt J, Strauss ME. 1990. Aspects of procedural memory are differentially impaired in Huntington’s disease. Arch Clin Neuropsychol 5:287-297. Canals JM, Checa N, Marco S, Akerud P, Michels A, Perez-Navarro E, Tolosa E, Arenas E, Alberch J. 2001. Expression of brain-derived neurotrophic factor in cortical neurons is regulated by striatal target area. J Neurosci 21:117-124. Cha JH, Kosinski CM, Kerner JA, Alsdorf SA, Mangiarini L, Davies SW, Penney JB, Bates GP, Young AB. 1998. Altered brain neurotransmitter receptors in transgenic mice expressing a portion of an abnormal human Huntington disease gene. Proc Natl Acad Sci U S A 95:64806485. Cicchetti F, Prensa L, Wu Y, Parent A. 2000. Chemical anatomy of striatal interneurons in normal individuals and in patients with Huntington’s disease. Brain Res Brain Res Rev 34:80-101. Coles CJ, Edmondson DE, Singer TP. 1979. Inactivation of succinate dehydrogenase by 3-nitropropionate. J Biol Chem 254:5161-5167. Coyle JT, Schwarcz R. 1976. Lesion of striatal neurons with kainic acid provides a model for Huntington’s chorea. Nature 263:244-246. Dawbarn D, De Quidt ME, Emson PC. 1985. Survival of basal ganglia neuropeptide Y-somatostatin neurons in Huntington’s disease. Brain Res 340:251-260. de Almeida LP, Ross CA, Zala D, Aebischer P, Déglon N. 2002. Lentiviral-mediated delivery of mutant huntingtin in the striatum of rats induces a selective neuropathology modulated by polyglutamine repeat

ILAR Journal

size, huntingtin expression levels, and protein length. J Neurosci 22: 3473-3483. Dedeoglu A, Kubilus JK, Yang L, Ferrante KL, Hersch SM, Beal MF, Ferrante RJ. 2003. Creatine therapy provides neuroprotection after onset of clinical symptoms in Huntington’s disease transgenic mice. J Neurochem 85:1359-1367. Emerich DF, Winn SR, Hantraye PM, Peschanski M, Chen EY, Chu Y, McDermott P, Baetge EE, Kordower JH. 1997. Protective effect of encapsulated cells producing neurotrophic factor CNTF in a monkey model of Huntington’s disease. Nature 386:395-399. Emerich DF, Bruhn S, Chu Y, Kordower JH. 1998. Cellular delivery of CNTF but not NT-4/5 prevents degeneration of striatal neurons in a rodent model of Huntington’s disease. Cell Transplant 7:213-225. Emerich DF, Thanos CG, Goddard M, Skinner SJ, Geany MS, Bell WJ, Bintz B, Schneider P, Chu Y, Babu RS, Borlongan CV, Boekelheide K, Hall S, Bryant B, Kordower JH. 2006. Extensive neuroprotection by choroid plexus transplants in excitotoxin lesioned monkeys. Neurobiol Dis 23:471-480. Ferrante RJ, Beal MF, Kowall NW, Richardson EP Jr, Martin JB. 1987a. Sparing of acetylcholinesterase-containing striatal neurons in Huntington’s disease. Brain Res 411:162-166. Ferrante RJ, Kowall NW, Beal MF, Martin JB, Bird ED, Richardson EP Jr. 1987b. Morphologic and histochemical characteristics of a spared subset of striatal neurons in Huntington’s disease. J Neuropathol Exp Neurol 46:12-27. Ferrante RJ, Kowall NW, Cipolloni PB, Storey E, Beal MF. 1993. Excitotoxin lesions in primates as a model for Huntington’s disease: Histopathologic and neurochemical characterization. Exp Neurol 119:4671. Foster AC, Miller LP, Oldendorf WH, Schwarcz R. 1984. Studies on the disposition of quinolinic acid after intracerebral or systemic administration in the rat. Exp Neurol 84:428-440. Furtado JC, Mazurek MF. 1996. Behavioral characterization of quinolinate-induced lesions of the medial striatum: Relevance for Huntington’s disease. Exp Neurol 138:158-168. Gardian G, Vecsei L. 2004. Huntington’s disease: Pathomechanism and therapeutic perspectives. J Neural Transm 111:1485-1494. Goldberg TE, Berman KF, Mohr E, Weinberger DR. 1990. Regional cerebral blood flow and cognitive function in Huntington’s disease and schizophrenia: A comparison of patients matched for performance on a prefrontal-type task. Arch Neurol 47:418-422. Gu M, Gash MT, Mann VM, Javoy-Agid F, Cooper JM, Schapira AH. 1996. Mitochondrial defect in Huntington’s disease caudate nucleus. Ann Neurol 39:385-389. Hahn-Barma V, Deweer B, Durr A, Dode C, Feingold J, Pillon B, Agid Y, Brice A, Dubois B. 1998. Are cognitive changes the first symptoms of Huntington’s disease? A study of gene carriers. J Neurol Neurosurg Psychiatry 64:172-177. Hargreaves KM, Pardridge WM. 1988. Neutral amino acid transport at the human blood-brain barrier. J Biol Chem 263:19392-19397. Harper SQ, Staber PD, He X, Eliason SL, Martins IH, Mao Q, Yang L, Kotin RM, Paulson HL, Davidson BL. 2005. RNA interference improves motor and neuropathological abnormalities in a Huntington’s disease mouse model. Proc Natl Acad Sci U S A 102:5820-5825. Hedreen JC, Peyser CE, Folstein SE, Ross CA. 1991. Neuronal loss in layers V and VI of cerebral cortex in Huntington’s disease. Neurosci Lett 133:257-261. Heindel WC, Butters N, Salmon DP. 1988. Impaired learning of a motor skill in patients with Huntington’s disease. Behav Neurosci 102:141147. Henderson LM, Hirsch HM. 1949. Quinolinic acid metabolism: Urinary excretion by the rat following tryptophan and 3-hydroxyanthranilic acid administration. J Biol Chem 181:667-675. Henderson LM, Ramasarma GB, Johnson BC. 1949. Quinolinic acid metabolism, urinary excretion by man and other mammals as affected by the ingestion of tryptophan. J Biol Chem 181:731-738. Hodgson JG, Agopyan N, Gutekunst CA, Leavitt BR, LePiane F, Singaraja R, Smith DJ, Bissada N, McCutcheon K, Nasir J, Jamot L, Li XJ,

Volume 48, Number 4

2007

Stevens ME, Rosemond E, Roder JC, Phillips AG, Rubin EM, Hersch SM, Hayden MR. 1999. A YAC mouse model for Huntington’s disease with full-length mutant huntingtin, cytoplasmic toxicity, and selective striatal neurodegeneration. Neuron 23:181-192. Huntington G. 2003. On chorea. J Neuropsychiatry Clin Neurosci 15:109112. Isacson O, Brundin P, Kelly PA, Gage FH, Bjorklund A. 1984. Functional neuronal replacement by grafted striatal neurones in the ibotenic acidlesioned rat striatum. Nature 311:458-460. Jackson GR, Salecker I, Dong X, Yao X, Arnheim N, Faber PW, MacDonald ME, Zipursky SL. 1998. Polyglutamine-expanded human huntingtin transgenes induce degeneration of Drosophila photoreceptor neurons. Neuron 21:633-642. Kasparova S, Sumbalova Z, Bystricky P, Kucharska J, Liptaj T, Mlynarik V, Gvozdjakova A. 2006. Effect of coenzyme Q10 and vitamin E on brain energy metabolism in the animal model of Huntington’s disease. Neurochem Int 48:93-99. Kendall AL, David F, Rayment G, Torres EM, Annett LE, Dunnett SB. 2000. The influence of excitotoxic basal ganglia lesions on motor performance in the common marmoset. Brain 123 (Pt 7):1442-1458. Kidd PM. 2005. Neurodegeneration from mitochondrial insufficiency: Nutrients, stem cells, growth factors, and prospects for brain rebuilding using integrative management. Altern Med Rev 10:268-293. Kordower JH, Isacson O, Leventhal L, Emerich DF. 2000. Cellular delivery of trophic factors for the treatment of Huntington’s disease: Is neuroprotection possible? Prog Brain Res 127:414-430. Lange KW, Sahakian BJ, Quinn NP, Marsden CD, Robbins TW. 1995. Comparison of executive and visuospatial memory function in Huntington’s disease and dementia of Alzheimer type matched for degree of dementia. J Neurol Neurosurg Psychiatry 58:598-606. Leavitt BR, Raamsdonk JM, Shehadeh J, Fernandes H, Murphy Z, Graham RK, Wellington CL, Raymond LA, Hayden MR. 2006. Wild-type huntingtin protects neurons from excitotoxicity. J Neurochem 96:11211129. Lee ST, Chu K, Park JE, Lee K, Kang L, Kim SU, Kim M. 2005. Intravenous administration of human neural stem cells induces functional recovery in Huntington’s disease rat model. Neurosci Res 52:243-249. Lin CH, Tallaksen-Greene S, Chien WM, Cearley JA, Jackson WS, Crouse AB, Ren S, Li XJ, Albin RL, Detloff PJ. 2001. Neurological abnormalities in a knock-in mouse model of Huntington’s disease. Hum Mol Genet 10:137-144. Lione LA, Carter RJ, Hunt MJ, Bates GP, Morton AJ, Dunnett SB. 1999. Selective discrimination learning impairments in mice expressing the human Huntington’s disease mutation. J Neurosci 19:10428-10437. Liu X, Luo X, Hu W. 1992. Studies on the epidemiology and etiology of moldy sugarcane poisoning in China. Biomed Environ Sci 5:161-177. Ludolph AC, He F, Spencer PS, Hammerstad J, Sabri M. 1991. 3-Nitropropionic acid-exogenous animal neurotoxin and possible human striatal toxin. Can J Neurol Sci 18:492-498. Luesse HG, Schiefer J, Spruenken A, Puls C, Block F, Kosinski CM. 2001. Evaluation of R6/2 HD transgenic mice for therapeutic studies in Huntington’s disease: Behavioral testing and impact of diabetes mellitus. Behav Brain Res 126:185-195. Machida Y, Okada T, Kurosawa M, Oyama F, Ozawa K, Nukina N. 2006. rAAV-mediated shRNA ameliorated neuropathology in Huntington disease model mouse. Biochem Biophys Res Commun 343:190-197. Mangiarini L, Sathasivam K, Seller M, Cozens B, Harper A, Hetherington C, Lawton M, Trottier Y, Lehrach H, Davies SW, Bates GP. 1996. Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87:493-506. McBride JL, During MJ, Wuu J, Chen EY, Leurgans SE, Kordower JH. 2003. Structural and functional neuroprotection in a rat model of Huntington’s disease by viral gene transfer of GDNF. Exp Neurol 181:213223. McBride JL, Behrstock SP, Chen EY, Jakel RJ, Siegel I, Svendsen CN, Kordower JH. 2004. Human neural stem cell transplants improve motor

371

function in a rat model of Huntington’s disease. J Comp Neurol 475: 211-219. McBride JL, Ramaswamy S, Gasmi M, Bartus RT, Herzog CD, Brandon EP, Zhou L, Pitzer MR, Berry-Kravis EM, Kordower JH. 2006. Viral delivery of glial cell line-derived neurotrophic factor improves behavior and protects striatal neurons in a mouse model of Huntington’s disease. Proc Natl Acad Sci U S A 103:9345-9350. McLin JP, Thompson LM, Steward O. 2006. Differential susceptibility to striatal neurodegeneration induced by quinolinic acid and kainate in inbred, outbred and hybrid mouse strains. Eur J Neurosci 24:31343140. Menalled LB, Sison JD, Dragatsis I, Zeitlin S, Chesselet MF. 2003. Time course of early motor and neuropathological anomalies in a knock-in mouse model of Huntington’s disease with 140 CAG repeats. J Comp Neurol 465:11-26. Ming L. 1995. Moldy sugarcane poisoning—a case report with a brief review. J Toxicol Clin Toxicol 33:363-367. Mohr E, Brouwers P, Claus JJ, Mann UM, Fedio P, Chase TN. 1991. Visuospatial cognition in Huntington’s disease. Mov Disord 6:127-132. Nasir J, Floresco SB, O’Kusky JR, Diewert VM, Richman JM, Zeisler J, Borowski A, Marth JD, Phillips AG, Hayden MR. 1995. Targeted disruption of the Huntington’s disease gene results in embryonic lethality and behavioral and morphological changes in heterozygotes. Cell 81:811-823. Naver B, Stub C, Moller M, Fenger K, Hansen AK, Hasholt L, Sorensen SA. 2003. Molecular and behavioral analysis of the R6/1 Huntington’s disease transgenic mouse. Neuroscience 122:1049-1057. Nguyen HP, Kobbe P, Rahne H, Worpel T, Jager B, Stephan M, Pabst R, Holzmann C, Riess O, Korr H, Kantor O, Petrasch-Parwez E, Wetzel R, Osmand A, von Horsten S. 2006. Behavioral abnormalities precede neuropathological markers in rats transgenic for Huntington’s disease. Hum Mol Genet 15:3177-3194. Nicholson LF, Faull RL, Waldvogel HJ, Dragunow M. 1995. GABA and GABAA receptor changes in the substantia nigra of the rat following quinolinic acid lesions in the striatum closely resemble Huntington’s disease. Neuroscience 66:507-521. Ouary S, Bizat N, Altairac S, Menetrat H, Mittoux V, Conde F, Hantraye P, Brouillet E. 2000. Major strain differences in response to chronic systemic administration of the mitochondrial toxin 3-nitropropionic acid in rats: Implications for neuroprotection studies. Neuroscience 97:521-530. Palfi S, Ferrante RJ, Brouillet E, Beal MF, Dolan R, Guyot MC, Peschanski M, Hantraye P. 1996. Chronic 3-nitropropionic acid treatment in baboons replicates the cognitive and motor deficits of Huntington’s disease. J Neurosci 16:3019-3025. Palfi S, Leventhal L, Goetz CG, Hantraye T, Roitberg BZ, Sramek J, Emborg M, Kordower JH. 2000. Delayed onset of progressive dystonia following subacute 3-nitropropionic acid treatment in Cebus apella monkeys. Mov Disord 15:524-530. Pang Z, Geddes JW. 1997. Mechanisms of cell death induced by the mitochondrial toxin 3-nitropropionic acid: Acute excitotoxic necrosis and delayed apoptosis. J Neurosci 17:3064-3073. Perez-Navarro E, Arenas E, Marco S, Alberch J. 1999. Intrastriatal grafting of a GDNF-producing cell line protects striatonigral neurons from quinolinic acid excitotoxicity in vivo. Eur J Neurosci 11:241-249. Perez-Navarro E, Akerud P, Marco S, Canals JM, Tolosa E, Arenas E, Alberch J. 2000a. Neurturin protects striatal projection neurons but not interneurons in a rat model of Huntington’s disease. Neuroscience 98: 89-96. Perez-Navarro E, Canudas AM, Akerund P, Alberch J, Arenas E. 2000b. Brain-derived neurotrophic factor, neurotrophin-3, and neurotrophin4/5 prevent the death of striatal projection neurons in a rodent model of Huntington’s disease. J Neurochem 75:2190-2199. Petersen A, Gil J, Maat-Schieman ML, Bjorkqvist M, Tanila H, Araujo IM, Smith R, Popovic N, Wierup N, Norlen P, Li JY, Roos RA, Sundler F, Mulder H, Brundin P. 2005. Orexin loss in Huntington’s disease. Hum Mol Genet 14:39-47. Pineda JR, Rubio N, Akerud P, Urban N, Badimon L, Arenas E, Alberch

372

J, Blanco J, Canals JM. 2007. Neuroprotection by GDNF-secreting stem cells in a Huntington’s disease model: Optical neuroimage tracking of brain-grafted cells. Gene Ther 14:118-128. Popovic N, Maingay M, Kirik D, Brundin P. 2005. Lentiviral gene delivery of GDNF into the striatum of R6/2 Huntington mice fails to attenuate behavioral and neuropathological changes. Exp Neurol 193:65-74. Portera-Cailliau C, Hedreen JC, Price DL, Koliatsos VE. 1995. Evidence for apoptotic cell death in Huntington disease and excitotoxic animal models. J Neurosci 15:3775-3787. Ramaswamy S, McBride JL, Zhou L, Berry-Kravis EM, Brandon EP, Herzog CD, Gasmi M, Bartus RT, Kordower JH. 2004. Cognitive deficits in the N171-82Q transgenic mouse model of Huntington’s disease. Cell Transplant. Special issue for the American Society for Neural Therapy and Repair. Ramaswamy S, McBride JL, Zhou L, Berry-Kravis EM, Brandon EP, Herzog CD, Gasmi M, Bartus RT, Kordower JH. 2006. AAV-neurturin (Cere-120) therapy in a transgenic mouse model of Huntington’s disease (SFN Abstract). Society for Neuroscience 472.11 (available at www.abstractsonline.com). Reddy PH, Williams M, Charles V, Garrett L, Pike-Buchanan L, Whetsell WO Jr, Miller G, Tagle DA. 1998. Behavioural abnormalities and selective neuronal loss in HD transgenic mice expressing mutated fulllength HD cDNA. Nat Genet 20:198-202. Reiner A, Albin RL, Anderson KD, D’Amato CJ, Penney JB, Young AB. 1988. Differential loss of striatal projection neurons in Huntington’s disease. Proc Natl Acad Sci U S A 85:5733-5737. Ribeiro CA, Grando V, Dutra Filho CS, Wannmacher CM, Wajner M. 2006. Evidence that quinolinic acid severely impairs energy metabolism through activation of NMDA receptors in striatum from developing rats. J Neurochem 99:1531-1542. Roberts RC, Ahn A, Swartz KJ, Beal MF, DiFiglia M. 1993. Intrastriatal injections of quinolinic acid or kainic acid: Differential patterns of cell survival and the effects of data analysis on outcome. Exp Neurol 124: 274-282. Roitberg BZ, Emborg ME, Sramek JG, Palfi S, Kordower JH. 2002. Behavioral and morphological comparison of two nonhuman primate models of Huntington’s disease. Neurosurgery 50:137-145. Ruddick JP, Evans AK, Nutt DJ, Lightman SL, Rook GA, Lowry CA. 2006. Tryptophan metabolism in the central nervous system: Medical implications. Expert Rev Mol Med 8:1-27. Schilling G, Becher MW, Sharp AH, Jinnah HA, Duan K, Kotzuk JA, Slunt HH, Ratovitski T, Cooper JK, Jenkins NA, Copeland NG, Price DL, Ross CA, Borchelt DR. 1999. Intranuclear inclusions and neuritic aggregates in transgenic mice expressing a mutant N-terminal fragment of huntingtin. Hum Mol Genet 8:397-407. Schilling G, Coonfield ML, Ross CA, Borchelt DR. 2001. Coenzyme Q10 and remacemide hydrochloride ameliorate motor deficits in a Huntington’s disease transgenic mouse model. Neurosci Lett 315:149-153. Schwarcz R, Okuno E, White RJ, Bird ED, Whetsell WO Jr. 1988. 3-Hydroxyanthranilate oxygenase activity is increased in the brains of Huntington disease victims. Proc Natl Acad Sci U S A 85:4079-4081. Selemon LD, Rajkowska G, Goldman-Rakic PS. 2004. Evidence for progression in frontal cortical pathology in late-stage Huntington’s disease. J Comp Neurol 468:190-204. Shear DA, Dong J, Haik-Creguer KL, Bazzett TJ, Albin RL, Dunbar GL. 1998. Chronic administration of quinolinic acid in the rat striatum causes spatial learning deficits in a radial arm water maze task. Exp Neurol 150:305-311. Shear DA, Haik KL, Dunbar GL. 2000. Creatine reduces 3-nitropropionicacid-induced cognitive and motor abnormalities in rats. Neuroreport 11:1833-1837. Shin JY, Fang ZH, Yu ZX, Wang CE, Li SH, Li XJ. 2005. Expression of mutant huntingtin in glial cells contributes to neuronal excitotoxicity. J Cell Biol 171:1001-1012. Singal SA, Briggs AP, Sydenstricker VP, Littlejohn JM. 1946. The effect of tryptophane on the urinary excretion of nicotinic acid in rats. J Biol Chem 166:573-579. Slow EJ, van Raamsdonk J, Rogers D, Coleman SH, Graham RK, Deng Y,

ILAR Journal

Oh R, Bissada N, Hossain SM, Yang YZ, Li XJ, Simpson EM, Gutekunst CA, Leavitt BR, Hayden MR. 2003. Selective striatal neuronal loss in a YAC128 mouse model of Huntington disease. Hum Mol Genet 12:1555-1567. Smith KM, Matson S, Matson WR, Cormier K, Del Signore SJ, Hagerty SW, Stack EC, Ryu H, Ferrante RJ. 2006. Dose ranging and efficacy study of high-dose coenzyme Q10 formulations in Huntington’s disease mice. Biochim Biophys Acta 1762:616-626. Stack EC, Kubilus JK, Smith K, Cormier K, Del Signore SJ, Guelin E, Ryu H, Hersch SM, Ferrante RJ. 2005. Chronology of behavioral symptoms and neuropathological sequela in R6/2 Huntington’s disease transgenic mice. J Comp Neurol 490:354-370. Stack EC, Smith KM, Ryu H, Cormier K, Chen M, Hagerty SW, Del Signore SJ, Cudkowicz ME, Friedlander RM, Ferrante RJ. 2006. Combination therapy using minocycline and coenzyme Q10 in R6/2 transgenic Huntington’s disease mice. Biochim Biophys Acta 1762:373380. Stone TW, Perkins MN. 1981. Quinolinic acid: A potent endogenous excitant at amino acid receptors in CNS. Eur J Pharmacol 72:411-412. Sun Z, Del Mar N, Meade C, Goldowitz D, Reiner A. 2002. Differential changes in striatal projection neurons in R6/2 transgenic mice for Huntington’s disease. Neurobiol Dis 11:369-385. Tabrizi SJ, Cleeter MW, Xuereb J, Taanman JW, Cooper JM, Schapira AH. 1999. Biochemical abnormalities and excitotoxicity in Huntington’s disease brain. Ann Neurol 45:25-32. Tatter SB, Galpern WR, Hoogeveen AT, Isacson O. 1995. Effects of striatal excitotoxicity on huntingtin-like immunoreactivity. Neuroreport 6: 1125-1129. The Huntington’s Disease Collaborative Research Group. 1993. A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell 72:971-983. Turmaine M, Raza A, Mahal A, Mangiarini L, Bates GP, Davies SW. 2000. Nonapoptotic neurodegeneration in a transgenic mouse model of Huntington’s disease. Proc Natl Acad Sci U S A 97:8093-8097. van Dellen A, Welch J, Dixon RM, Cordery P, York D, Styles P, Blakemore C, Hannan AJ. 2000. N-acetylaspartate and DARPP-32 levels decrease in the corpus striatum of Huntington’s disease mice. Neuroreport 11:3751-3757.

Volume 48, Number 4

2007

Van Raamsdonk JM, Pearson J, Slow EJ, Hossain SM, Leavitt BR, Hayden MR. 2005. Cognitive dysfunction precedes neuropathology and motor abnormalities in the YAC128 mouse model of Huntington’s disease. J Neurosci 25:4169-4180. Vazey EM, Chen K, Hughes SM, Connor B. 2006. Transplanted adult neural progenitor cells survive, differentiate and reduce motor function impairment in a rodent model of Huntington’s disease. Exp Neurol 199:384-396. von Horsten S, Schmitt I, Nguyen HP, Holzmann C, Schmidt T, Walther T, Bader M, Pabst R, Kobbe P, Krotova J, Stiller D, Kask A, Vaarmann A, Rathke-Hartlieb S, Schulz JB, Grasshoff U, Bauer I, Vieira-Saecker AM, Paul M, Jones L, Lindenberg KS, Landwehrmeyer B, Bauer A, Li XJ, Riess O. 2003. Transgenic rat model of Huntington’s disease. Hum Mol Genet 12:617-624. Vonsattel JP, Myers RH, Stevens TJ, Ferrante RJ, Bird ED, Richardson EP Jr. 1985. Neuropathological classification of Huntington’s disease. J Neuropathol Exp Neurol 44:559-577. Wang YL, Liu W, Wada E, Murata M, Wada K, Kanazawa I. 2005. Clinico-pathological rescue of a model mouse of Huntington’s disease by siRNA. Neurosci Res 53:241-249. Wheeler VC, Auerbach W, White JK, Srinidhi J, Auerbach A, Ryan A, Duyao MP, Vrbanac V, Weaver M, Gusella JF, Joyner AL, MacDonald ME. 1999. Length-dependent gametic CAG repeat instability in the Huntington’s disease knock-in mouse. Hum Mol Genet 8:115-122. Wheeler VC, White JK, Gutekunst CA, Vrbanac V, Weaver M, Li XJ, Li SH, Yi H, Vonsattel JP, Gusella JF, Hersch S, Auerbach W, Joyner AL, MacDonald ME. 2000. Long glutamine tracts cause nuclear localization of a novel form of huntingtin in medium spiny striatal neurons in HdhQ92 and HdhQ111 knock-in mice. Hum Mol Genet 9:503-513. Yamamoto A, Lucas JJ, Hen R. 2000. Reversal of neuropathology and motor dysfunction in a conditional model of Huntington’s disease. Cell 101:57-66. Yang L, Calingasan NY, Chen J, Ley JJ, Becker DA, Beal MF. 2005. A novel azulenyl nitrone antioxidant protects against MPTP and 3-nitropropionic acid neurotoxicities. Exp Neurol 191:86-93. Yu ZX, Li SH, Evans J, Pillarisetti A, Li H, Li XJ. 2003. Mutant huntingtin causes context-dependent neurodegeneration in mice with Huntington’s disease. J Neurosci 23:2193-2202.

373