Overview

Once considered a relatively rare disorder, Huntington’s disease (HD) has been historically important and continues to be at the frontier of human neurological diseases. The search for the HD gene has been at the forefront of the revolution in molecular genetics. Similarly, the pathological features of HD have stimulated the idea that neurotransmitters might themselves play a role in the etiology of neurologic diseases. Some of the first evidence that overactivity of the glutamate neurotransmission system could lead to neuronal death--a process termed excitotoxicity--followed from observations that administration of glutamate receptor agonists reproduces certain neuropathologic features of HD. The discovery of the defect in the mutant huntingtin gene places HD into a novel family of neurologic diseases, the trinucleotide repeat disorders. Most recently, the observation of abnormal properties of mutant huntingtin leading to formation of protein aggregates demonstrates that HD--like Alzheimer’s disease, prion disorders, and Parkinson’s disease--may be primarily a disease of abnormal protein conformation.

Historical Aspects

In 1872, the physician George Huntington described the neurologic disorder that has come to bear his name (Huntington, 1872; Durbach & Hayden, 1993). This disorder was originally called Huntington’s Chorea, from the Greek word for dance. Chorea refers to the dance-like adventitious movements that are the hallmark of the disorder. However, since chorea is not a universal feature, especially in the juvenile forms of the disease, the term Huntington’s disease is now used. In his description of this familial disorder, Huntington fittingly drew upon the observations of his father and grandfather, both of whom were physicians. It was over a hundred years before major advancements beyond Huntington’s prescient original description were made. Huntington’s original report noted the autosomal dominant inheritance, adult onset, the constellation of emotional, cognitive and motor symptoms, and the inexorably progressive and ultimately fatal nature of the disease.

The HD patients described by Huntington probably had ancestors in England, but the next chapter was written in Venezuela. In 1955, a young medical doctor named Amerigo Negrette arrived to be the local doctor in San Luis, a small fishing village just outside Maracaibo, Venezuela. Negrette was shocked to find that many of the villagers were staggering around as if they were drunk, even early in the day. He quickly learned that the villagers were not drunk; they had a familial disorder known locally as El Mal de San Vito (St. Vitus’ Dance). Negrette was perspicacious enough to conclude that el Mal de San Vito was in fact Huntington’s disease. Negrette, realizing that an important clue lay right within the village, set out to study and examine carefully the residents of San Luis (Negrette, 1958).

In 1972, a symposium to commemorate the one hundred-year anniversary of Huntington’s paper was held in Columbus, Ohio (Barbeau et al., 1973). Physicians and scientists from around the world assembled to discuss what they considered to be a rare disorder. Ramon Avila-Giron, a colleague of Negrette’s, stunned the audience by showing a movie of a Venezuelan village filled with HD patients (Avila-Giron, 1973). Never had such numbers of HD patients been described.

In the audience was Nancy Wexler, whose own mother died of HD. Wexler, a Ph.D. in Psychology, was at that point at the National Institute of Neurological and Communicative Disorders and Stroke. In 1979, she led the first expedition to the city of Maracaibo in the state of Zulia, Venezuela. It was around the shores of Lake Maracaibo that the members of the family studied by Negrette lived. Wexler, along with Thomas Chase of the National Institutes of Health, surveyed the area and interviewed numerous family members. A collaboration between the NIH and the University of Zulia was hatched, and every spring since 1981, a team of scientists and physicians has gone to study this, the world’s largest known population of HD patients (Young et al., 1986c; Penney et al., 1990). It is through DNA samples collected from Venezuelan patients that the first marker for the HD gene was found (Gusella et al., 1983), leading ultimately to the discovery of the gene itself (Huntington's Disease Collaborative Research Group, 1993). The U.S.-Venezuela Collaborative Huntington’s Project continues to make yearly expeditions to the Maracaibo region, gathering important longitudinal and prospective data.

Clinical Features

Huntington’s disease is typically an adult onset disorder characterized by insidious onset of both neurologic and psychiatric symptoms (Ranen et al., 1993; Harper, 1996). In the United States, approximately 25,000 persons are affected by HD (about 10 per 100,000 population) and approximately 150,000 persons are at 50% risk for the disease by virtue of having an affected parent (Conneally, 1984). Symptoms usually begin in the mid-thirties to mid-forties, although disease onset can range from as young as 2 years or as old as 80 years. Initial symptoms include personality change and the gradual appearance of small involuntary movements (Young et al., 1986b; Harper, 1996). Symptoms progress, with chorea becoming more obvious and incapacitating. Over years, motor symptoms worsen such that walking becomes more difficult, as do speaking and eating. Weight loss is common, partially due to the extra energy required for adventitious movements but also to increased resting basal energy expenditure. Most HD patients eventually succumb to aspiration pneumonia, because of swallowing difficulties.

About 10% of cases start before the age of 20. The juvenile form (Westphal variant) is more parkinsonian in nature. Rather than chorea, the prominent features are bradykinesia, rigidity, and tremor (Table 1). Seizures can be present in juvenile onset. Juvenile onset HD usually results from paternal transmission (see below). Individuals who develop symptoms before age 10, more than 90% have an affected father (Hayden, 1981; Conneally, 1984; Folstein, 1989). This tendency for anticipation—younger age on onset in successive generations—is especially pronounced in cases of paternal transmission. Interestingly, both the juvenile onset and adult onset phenotypes can be present within the same family (Young et al., 1986b).

New mutations have been described, usually from a parent carrying in “intermediate allele” (see below). Homozygotes for the HD gene have also been described (Wexler et al., 1991). The clinical symptomatology manifested by these very few patients appears to be no worse than that manifested by patients carrying only a single HD gene. Thus, HD appears to be a truly dominant disease.

Pathology

HD is a neurodegenerative disease, meaning that there is progressive dysfunction and destruction of a brain that has undergone normal development. Neuropathologically, HD is characterized by atrophy of the basal ganglia, specifically the striatum (caudate-putamen) (Vonsattel et al., 1985). The basal ganglia are a set of subcortical gray matter structures, which are involved in various aspects of motor control, cognition, and sensory pathways (Graybiel, 1990) (Figure 1). Pathologic changes have also been described in cortex, thalamus, and subthalamic nucleus (Hedreen et al., 1991). Atrophy and gliosis of the caudate nucleus and putamen is progressive and marked. Atrophy and gliosis also occurs in other basal ganglia structures, including the lateral and medial globus pallidus and the substantia nigra pars reticulatum. The apparent increase in neuronal density within the pallidum may in fact represent loss of inputs from the striatopallidal projections. In juvenile onset cases, cell loss tends to be more pronounced, and can also be seen in the cerebellum.

Within the striatum, HD differentially affects subpopulations of neurons, with projection neurons preferentially being lost rather than interneurons (Albin et al., 1989; DiFiglia, 1990). The g-aminobutyric acid (GABA)-containing medium spiny neurons which are the output projection neurons comprise approximately 90% of striatal neurons. There are multiple type of interneurons, including 1) large acetylcholine-containing aspiny neurons, 2) medium somatostatin/neuropeptide Y/diaphorase/nitric oxide synthase-positive aspiny neurons, and 3) medium GABA-containing/parvalbumin-positive aspiny neurons (Kawaguchi et al., 1995). HD seems to spare the interneurons while devastating the projection neurons (Ferrante et al., 1987a; Ferrante et al., 1987b). Consistent with the finding of loss of projection neurons was the early finding that GABA levels were markedly reduced in the caudate-putamen of HD patients (Perry et al., 1973).

The projection neurons are themselves divided into two classes. Both types contain GABA as a primary neurotransmitter, but they differ with respect to (i) which neuropeptide they employ as a co-transmitter and (ii) the synaptic targets. The substance P/dynorphin/GABA-containing neurons project to the medial globus pallidus and the substantia nigra and thus constitute the first portion of the direct pathway (Figure 1) (Albin et al., 1989). Activation of the direct pathway tends to increase movement. That is, substance P/dynorphin/GABA-containing striatal neurons, when activated, will inhibit the medial globus pallidus, resulting in disinhibition of the VA and VL thalamic nuclei, which will in turn excite the cortex. In contrast, the enkephalin/GABA-containing neurons project to the lateral globus pallidus and thus constitute the first portion of the indirect pathway. Activation of the indirect pathway tend to decrease movement. That is, enkephalin/GABA-containing striatal neurons, when activated, will inhibit the lateral globus pallidus, resulting in disinhibition of the medial globus pallidus, which then inhibits VA/VL of thalamus, ultimately resulting in reduced cortical excitation. Another level of striatal organization is a functional one based initially on acetylcholinesterase (AChE) staining patterns. AChE staining reveals a spotted pattern in the caudate-putamen, in which AChE-pale patches (“striosomes”) are found on a background (“matrix”) of high AChE staining (Graybiel, 1990). Hedreen and Folstein have suggested that striatal degeneration in HD proceeds in a dorsal to ventral fashion, with striosomal compartments affected before matrix compartments (Hedreen & Folstein, 1995).

Of the two populations of striatal projection neurons, the neurons of the indirect pathway (i.e. enkephalin/GABA-containing neurons) are affected first, thus providing an anatomical substrate for the increased movement which is the hallmark of HD (Reiner et al., 1988; Albin et al., 1992). Consistent with this observation is the fact that in early grade HD cases, markers for these striatal neurons are decreased, including dopamine D2 receptors, adenosine A2a receptors and enkephalin. In early HD, thus, the indirect pathway is predominantly disrupted, and basal ganglia circuitry models predict an overall increase in movement, manifested as chorea and ballism (Albin et al., 1989). In later stages of adult HD, both populations of striatal projection neurons are affected, with concomitant loss of markers of the direct pathway (i.e. substance P/dynorphin/GABA-containing neurons), including dopamine D1 receptors and substance P (Reiner et al., 1988; Richfield et al., 1991). The functional correlates of degeneration of both the direct and indirect pathways is a rigid bradykinetic state (Albin et al., 1989). In juvenile HD cases which resemble Parkinson’s disease, degeneration of both direct and indirect pathway striatal neurons are seen (Albin et al., 1988; Albin et al., 1990).

The mode by which neurons die in HD is unknown. One theory is that neuronal death in neurons occurs through apoptosis, that is, activation of a programmed cell death pathway. DNA fragmentation characteristic of apoptosis has been described in HD brain (Dragunow et al., 1995; Portera-Cailliau et al., 1995; Thomas et al., 1995). However, it is not yet clear whether apoptosis is the primary pathologic event, or merely the final mode of exit for neurons which have been damaged by some other process.

Certain symptoms of HD are likely explained by the anatomy of neurodegeneration. Dysfunction of the neurons of the indirect pathway, as occurs in early HD, is predicted to produce chorea and ballism. Degeneration of both the indirect and direct pathways is predicted to account for the bradykinetic rigid phenotype observed in late stage HD. Dementia and personality change likely result from cortical damage (de la Monte et al., 1988; Cudkowicz & Kowall, 1990; Hedreen et al., 1991). In sum, the pathology of HD is strikingly specific, raising several critical questions. Why is the striatum predisposed to neurodegeneration? Why are certain populations of striatal neurons selectively targeted or spared in HD?

Theories of HD Pathogenesis: Excitotoxicity and Mitochondrial Dysfunction

Prior to the discovery of the Huntington’s disease (HD) gene, the leading hypotheses concerning the pathogenesis of HD implicated either excitotoxicity or energetic dysfunction. Evidence for both theories has been demonstrated in both human and animal model studies of HD (Beal et al., 1986; DiFiglia, 1990; Albin & Greenamyre, 1992). Discovery of the HD gene in 1993 did not reveal the pathogenetic mechanism of HD (Huntington's Disease Collaborative Research Group, 1993). The novel protein huntingtin has no clear relationship to excitatory amino acid neurotransmission, nor to cellular, and particularly mitochondrial, energetics.

Excitotoxicity is the process in which neuronal cells die as a result of excessive excitatory amino acid neurotransmission. Excitatory amino acids, especially glutamate, have been postulated to play a role in the pathogenesis of HD because intrastriatal injections of excitatory amino acid receptor agonists, particularly agonists acting at the N-methyl-D-aspartate subtype of glutamate receptor, reproduce the neuropathological features of HD (Beal et al., 1986; DiFiglia, 1990; Albin & Greenamyre, 1992). Glutamate has been postulated to kill neurons in a number of neurological disorders, including hypoxia-ischemia, head trauma, epilepsy, schizophrenia, and neurodegenerative disorders such as Alzheimer’s disease, Parkinson’s disease and amyotrophic lateral sclerosis (Fagg et al., 1986; Robinson & Coyle, 1987; Choi & Rothman, 1990; Klockgether & Turski, 1993; Patel & Zinkand, 1994; Bittigau & Ikonomidou, 1997; Goff & Wine, 1997; Choi et al., 1998; Kornbuber & Wiltfang, 1998). Postmortem analyses of HD brains demonstrate reduced glutamate receptors, although this loss may represent generalized neuronal loss (London et al., 1981; Young et al., 1988; Dure et al., 1991).

Mitochondrial dysfunction has also been implicated as a pathologic mechanism in HD, potentially rendering cells vulnerable to normal ambient levels of extracellular glutamate (Beal et al., 1986; Albin & Greenamyre, 1992). Positron emission tomography studies have demonstrated reduced striatal glucose metabolism in HD patients (Kuhl et al., 1985; Young et al., 1986a; Mazziotta et al., 1987). MRI spectroscopy shows increased levels of striatal glutamate/glutamine and lactate in HD patients suggesting that glutamatergic function and abnormalities in energy metabolism may combine to produce pathology (Jenkins et al., 1993; Taylor-Robinson et al., 1996; Browne et al., 1997; Koroshetz et al., 1997). Defects in mitochondrial metabolism have been demonstrated in HD brain tissue, especially in complex II-III activity (Brennan et al., 1985; Butterworth et al., 1985; Mann et al., 1990; Parker et al., 1990; Gu et al., 1996; Browne et al., 1997). Further, intrastriatal injection of the complex II inhibitor malonate produces lesions reminiscent of HD (Beal et al., 1993a; Henshaw et al., 1994; Greene & Greenamyre, 1995). Systemic administration of another complex II inhibitor, 3-nitropropionic acid (3NP), produces strikingly focal striatal lesions in rodents and primates, also reminiscent of HD striatal pathology (Beal et al., 1993b; Brouillet et al., 1993; Wuellner et al., 1994; Borlongan et al., 1995; Palfi et al., 1996). Reduced complex IV activity has also been described in HD brain (Gu et al., 1996; Browne et al., 1997). Mitochondrial aconitase, a metalloprotein involved in the tricarboxylic acid cycle whose function is inhibited by free radicals, has been shown to have decreased activity in HD caudate and putamen (Schapira et al., 1997). Phosphocreatine (PCr) is thought to be a high energy storage molecule, and ratios of PCr/Pi are decreased in muscle of HD patients (Koroshetz et al., 1997). Finally, administration of coenzyme Q₁₀, an essential cofactor of the electron transport chain, lowers elevated cortical lactate levels in HD patients back to levels seen in normal controls (Koroshetz et al., 1997).

Both excitotoxicity and mitochondrial dysfunction provide insights into potential mechanisms of cell death in HD (Table 2). Yet the novel protein huntingtin is not clearly related to either. Huntingtin does interact with glyceraldehye-3-phosphate dehydrogenase (GAPDH), an essential enzyme in glycolysis, although the significance of this interaction is as yet unclear (Burke et al., 1996a). GAPDH activity in HD basal ganglia was not found to be significantly different from controls (Browne et al., 1997; Schapira et al., 1997).

Huntington’s Disease Gene

Recombinant DNA techniques became available in the late 1970’s, and their power was brought to bear on the search for the HD gene (Wexler et al., 1991). Such technology had never been used to localize a human disease gene. Using anonymous DNA markers spread over the human genome, David Housman at the Massachusetts Institute of Technology and James Gusella at Massachusetts General Hospital used DNA from the large Venezuelan family to localize the HD gene to the long arm of chromosome 4. This remarkable discovery was a triumph, heralding the power of recombinant DNA technology in the search for disease genes. Discovery of an HD gene marker raised hopes that identification of the gene would be soon in coming.

It took ten years following the discovery of a genetic marker for HD that the gene itself was finally found. In 1993, the Huntington’s Disease Collaborative Research Group, a consortium of gene hunter laboratories from around the world organized under the auspices of the Hereditary Disease Foundation, isolated a gene termed IT-15 (“interesting transcript 15”) (Huntington's Disease Collaborative Research Group, 1993). Once IT-15 was found, it was obvious why the search for the gene had taken so long.

The mutation in the IT-15 gene was not one of the usual genetic mutations that are observed in human diseases: point mutation, deletion, duplication, or missense mutation. Rather, the defect in IT-15 resided in an unstable region of the gene, unstable in that a trinucleotide repeat motif present in normal alleles was expanded in the alleles of HD patients. CAG is the codon for glutamine, and this trinucleotide repeat gives rise to a polyglutamine moiety within the huntingtin protein. Normal huntingtin alleles contain from 6 to 35 CAG repeats, giving rise to 6 to 35 glutamines in the mature protein and variation occurs in the normal human population. Patients with Huntington’s disease invariably have alleles with greater than 35 repeats. While repeats greater than 40 invariably give rise to Huntington’s disease, there is a “gray area,” between 35-39 repeats, where some uncertainty exists. Some patients with up to 39 repeats have lived into their 70’s without developing overt signs of HD, although age of onset as late as 80 years has been described.

All new mutations have been shown to arise from “intermediate alleles” (IA) in one parent, with CAG ranging from between 29-35, passing through the paternal germline to produce repeat sizes of greater than 36. (Goldberg et al., 1993; Myers et al., 1993; Goldberg et al., 1995). Mosaicism in individual sperm DNA relative to somatic cell DNA obtained from lymphoblasts has been established in HD (Duyao et al., 1993; MacDonald et al., 1993; Telenius et al., 1994). While repeat numbers tend to be stable through successive generations, there are variations in the CAG repeat number. Single sperm PCR reveals that individual sperm have a distribution of repeat sizes, with some sperm having fewer CAG repeats than the paternal somatic cells but the majority of sperm having increased numbers of repeats (Zuhlke et al., 1993; Telenius et al., 1994; Leeflang et al., 1995).

Age of onset correlates negatively with repeat length, although the correlation is strongest for high CAG repeat numbers (Snell et al., 1993; Telenius et al., 1993). That is, although repeat numbers of greater than 70 invariably produce juvenile onset of HD, more common repeat numbers, for example, in the 40 to 45 range, have a varied age of onset (Duyao et al., 1993; Harper, 1996). Patients with very late onset tend to have repeats of 36 – 38, in the low abnormal range (James et al., 1994). Neuropathologic grade also varies with CAG number, with the most damage seen in brains with the highest CAG repeats (Penney et al., 1997).

Identification of the HD gene also explained how a single gene could have such diverse phenotypes. For example, within the large Venezuelan HD population, there is tremendous variation in age of onset of affected members (Young et al., 1986b; Penney et al., 1990). Further, within the same pedigree, different family members can have either typical adult onset HD or the atypical juvenile form. Much, but not all, of the variability of age of onset is accounted for by CAG repeat length, making it likely that other genetic modifiers play an important role in determining when symptoms begin.

HD thus belongs to a novel family of neurologic diseases, the trinucleotide repeat disorders (Paulson & Fischbeck, 1996). Certain diseases such as myotonic dystrophy, fragile X syndrome, and Friedreich’s ataxia are characterized by trinucleotide repeat expansions in non-coding regions of the gene. In contrast, HD belongs to a family of disorders characterized by expansion of a CAG trinucleotide motif within the coding region of the protein. CAG is the codon for glutamine, and thus the expansion regions code for polyglutamine moieties within the various proteins. The CAG repeat disorders, or polyglutamine disorders, display common features including dominant inheritance, adult onset, threshold levels of approximately 37 repeats, and progressive neurodegeneration (Table 3).

Huntingtin

IT-15 encodes a novel protein called huntingtin (Huntington's Disease Collaborative Research Group, 1993). The huntingtin gene comprises 67 exons, encoding a 3144 amino acid protein with an expected molecular weight of 348 kilodaltons. Expectations were that the distribution of huntingtin would mirror the characteristic distribution of neuropathological damage. However, huntingtin mRNA is expressed in almost all tissues of the body and homogenously throughout the brain (Li et al., 1993; Strong et al., 1993; Ambrose et al., 1994; Landwehrmeyer et al., 1995). No differences in IT-15 distribution is seen between HD and control cases (Li et al., 1993; Strong et al., 1993). Immunohistochemical studies confirm the widespread neuronal distribution of the huntingtin protein (DiFiglia et al., 1995; Gutekunst et al., 1995a; Ferrante et al., 1997; Kosinski et al., 1997).

The normal function of huntingtin is not completely known, although it has been implicated in membrane recycling. Huntingtin is transported by fast axonal transport (Block-Galarza et al., 1997), and may be involved in membrane recycling (DiFiglia et al., 1995). Huntingtin plays some important role in early development, since mice with targeted knockout of the huntingtin gene are embryonic lethal (Nasir et al., 1995; Zeitlin et al., 1995; White et al., 1997). Hemizygous knockout mice develop normally. Humans with Wolf-Hirshhorn syndrome, in which patients have a partial deletion of chromosome 4 and therefore have only one copy of the huntingtin gene, do not have the same neurologic disease as HD (Gusella et al., 1985). Therefore, having a single copy of a normal huntingtin allele does not produce the HD phenotype and rescues the embryonic lethality associated with having no copies of a huntingtin allele.

Mutant huntingtin is thought to have a gain of function, although the novel function is still unknown. The mutant protein is expressed in HD brain (Aronin et al., 1995; Gutekunst et al., 1995b; Persichetti et al., 1995). Mutant huntingtin possesses at least certain of the properties of wild-type huntingtin, in that mutant huntingtin can rescue the embryonic lethal phenotype seen in huntingtin-null knockout mice (Hodgson et al., 1996; Dragatsis et al., 1998).

Expanded length huntingtin interacts more strongly with a series of proteins more strongly than normal length huntingtin, providing a mechanism for the gain of function of the mutant gene. Most of these proteins have been isolated using the yeast two-hybrid system. Proteins interacting more strongly with mutant forms of huntingtin include: huntingtin-associated protein-1 (HAP-1) (Li et al., 1996), calmodulin (Bao et al., 1996), cystathionine-b-synthase (Boutell et al., 1998), the enzyme glyceraldehyde-3-phosphate dehydrogenase (Burke et al., 1996b), huntingtin-interacting protein (HIP-1) (Wanker et al., 1997), and SH3GL3 (Sittler et al., 1998). A series of proteins termed the HYP’s (Faber et al., 1998) have been found, all of which interact with the N-terminal end of the huntingtin protein, as opposed to the middle portion of the protein or the C-terminal. Huntingtin-interacting proteins remain an interesting area of inquiry, especially for those interactors which have higher affinity for mutant length huntingtin.

Transgenic Mouse Models

Descriptive studies based on postmortem human HD brain samples have been confounded by the profound cell loss which is characteristic of late HD. Exciting transgenic mouse models of HD have now been created, some of which have symptoms which resemble those seen in human HD (Mangiarini et al., 1996; Reddy et al., 1998b). Transgenic mice which appear normal at birth and later go on to develop HD-like symptoms thus offer an opportunity to examine which pathologic changes occur first and are not simply the result of cell loss. Ultimately, transgenic animals hold the promise of revealing those processes that cause neurons to die in HD. Interestingly, symptoms in transgenic mice occur long before neuronal cell loss is apparent, suggesting that a period of neuronal dysfunction precedes overt neuronal death. Several of the mouse lines develop abnormal neuronal intranuclear inclusions (NII) prior to the onset of symptoms (Davies et al., 1997). Indeed, nuclear inclusions have been seen in human HD (Roizin et al., 1976; DiFiglia et al., 1997; Becher et al., 1998), and in other human CAG triplicate diseases (Paulson et al., 1997; Skinner et al., 1997). Thus, the presence of NII may represent a unifying pathologic mechanism. The striking discovery of inclusions in both transgenic HD mice and human HD brain are undoubtedly a major step in understanding the mechanism of HD. However, the mechanism by which NII might produce clinical symptoms in the absence of overt neuronal cell loss is as yet unexplained.

One line of transgenic mice (R6/2 line) has been shown to have heightened susceptibility to 3NP-induced excitotoxic insult (Bogdanov et al., 1998), and also to manifest neurotransmitter abnormalities reminiscent of human HD (Cha et al., 1998). Neurotransmitter reductions occur at the level of receptor mRNA and precede the onset of clinical symptoms. Interestingly, positron emission tomography (PET) studies of presymptomatic HD gene-positive neurologically normal patients have demonstrated decreased dopamine receptors (Antonini et al., 1996; Weeks et al., 1996), and mRNA molecules for dopamine receptors and other neurotransmitters are decreased in early grade cases (Augood et al., 1996; Augood et al., 1997). Thus, transgenic mice offer the promise of being able to recapitulate the earliest neuropathological events in HD.

Polyglutamines and Aggregates

The polyglutamine moiety which is expanded in mutant forms of the huntingtin protein gives rise to the neuropathologic changes observed in HD. A transgenic mouse model expressing only exon 1 of the huntingtin gene (the portion containing the polyglutamine moiety) develops an abnormal neurologic phenotype (Mangiarini et al., 1996). Other transgenic mouse models expressing CAG repeat disease genes develop similar disease phenotypes, including transgenic mice for HD (Reddy et al., 1998a), DRPLA, spinocerebellar ataxia-1 (Burright et al., 1995), Machado-Joseph disease (Ikeda et al., 1996). In addition, one transgenic mouse model expressing exogenously placed CAG repeats placed within the HPRT locus also develops a neurologic phenotype reminiscent of HD (Ordway et al., 1997). Thus, it appears that expression of expanded polyglutamines, either within the context of known disease genes or ectopically expressed in an unrelated gene, can lead to neuronal dysfunction and abnormal behavior.

One of the striking findings which emerged out of the findings with transgenic animals is the observation of novel nuclear structures (Ross, 1997; Davies et al., 1998). Huntingtin is normally a cytoplasmic protein. Neuronal intranuclear inclusions were first observed in transgenic animals by Davies et al. (Davies et al., 1997). This remarkable finding prompted reexamination of human biopsy (Roizin et al., 1979) and necropsy (DiFiglia et al., 1997; Becher et al., 1998) material, and abnormal huntingtin- and ubiquitin-positive aggregates were also found. Although abnormal aggregates are a striking feature of human HD and transgenic mouse HD models, recent studies have raised controversy as to the importance of inclusions. In cell culture models, the presence of inclusions was not correlated with neuronal death (Saudou et al., 1998; Kim et al., 1999), suggesting that nuclear inclusions may themselves be an epiphenomenon, rather than the primary causal agent, of neuronal death.

Polyglutamines are found within proteins, including transcription factors (Gerber et al., 1994; Karlin & Burge, 1996). Several theories have emerged concerning the gain-of-function which emerges when the length of the polyglutamine moiety extends into the pathologic range. One theory proposes that polyglutamine moieties serve as a substrate for the enzyme transglutaminase (Kahlem et al., 1996; Lorand, 1996). Transglutaminase has been implicated in the role of cellular differentiation etc. In addition, transglutaminase is able to catalyze aggregation of the huntingtin protein, especially in the expanded form (Kahlem et al., 1998).

Another theory concerning the role of polyglutamines evolves form the observation that polyglutamine chains could theoretically form polar zippers (Perutz et al., 1994; Perutz, 1995; Perutz, 1996). Indeed, in vitro studies demonstrate convincingly that exon 1 of huntingtin protein can spontaneously aggregate in a fashion dependent on time, concentration and CAG repeat length (Scherzinger et al., 1997). the resulting fibrils have some properties of amyloid proteins, and are insoluble.

Therapeutics

There are currently no effective therapies for preventing the onset or slowing the progression of HD (Ranen et al., 1993). Current therapies are symptomatic, and include the use of neuroleptics to decrease chorea, and the use of psychotropic medications to address depression, obsessive compulsive symptoms, or psychosis. In addition, speech therapy and physical therapy are useful in addressing the swallowing and walking difficulties that many HD patients experience (Ranen et al., 1993).

The pathologic mechanisms underlying HD are not yet completely understood; effective therapeutics depends on a more clear elucidation of pathogenic pathways (Shoulson, 1998). The Huntington Study Group, a multicenter academic consortium of Huntington’s disease clinicians, is currently operating a clinical trial of two compounds, both alone and in combination. Remacemide is a compound which acts as a NMDA type glutamate receptor antagonist (Kieburtz et al., 1996), and coenzyme Q₁₀ is a compound which is directed towards maintaining the mitochondrial membrane potential (Feigin et al., 1996). Further therapies will no doubt arise as the molecular pathogenesis HD is further delineated.

published 2000