. beta-amyloid peptides enhance alpha-synuclein accumulation and neuronal deficits in a transgenic mouse model linking Alzheimer's disease and Parkinson's disease. Proc Natl Acad Sci U S A. 2001 Oct 9;98(21):12245-50. PubMed.

Recommends

Please login to recommend the paper.

Comments

Make a Comment

To make a comment you must login or register.

Comments on Primary Papers and News

  1. α-Synuclein and β-amyloid: A Linking of Partners to Model Disease

    Creating transgenic models of neurodegenerative disease has yielded mixed results. The transgenic models of Huntington's chorea and amyotrophic lateral sclerosis have yielded animal models showing a striking resemblance to their human counterparts (Davies et al., 1997; Gurney et al., 1994). On the other hand, developing transgenic animal models of Alzheimer's disease (AD) and synucleinopathies, such as Parkinson's disease or diffuse Lewy body disease, has proven to be difficult because mice that are transgenic for any single genes incompletely model these diseases. The steady march of progress has benefited the field, because the technique of combining transgenes appears has produced much more promising results. Most recently, a paper by Masliah and colleagues describes a mouse that carries transgenes for both amyloid precursor protein (APP) and a-synuclein, and that shows a striking resemblance to diffuse Lewy body disease (DLBD) (Masliah et al., 2001).

    The evolution of transgenic models of Alzheimer's disease serves as a useful reference for understanding the evolution of research in synucleinopathies. Initial attempts to model Alzheimer's disease by overexpressing amyloid precursor protein (APP) were unsuccessful because the animals did not develop neuritic plaques and neurofibrillary tangles. Partial success in modeling AD was achieved by using APP constructs that contained mutations associated with Alzheimer's disease. The PDAPP, Tg2576 and TgCRND8 mouse all contain mutant forms of APP that drive production of Aβ40 or Aβ42, develop neuritic plaques and, therefore, mimic the neuritic aspects of Alzheimer's disease (forming neuritic plaques) (Chishti et al., 2001; Games et al., 1995; Hsiao et al., 1996). Combining mutant APP with mutant presenilin accelerated the speed of development of neuritic plaques. The Alzheimer's-related presenilin mutations increase production of Aβ42, and mice that are transgenic for both mutant presenilin and mutant APP produce over 40 times more Aβ42, and develop neuritic plaques at a much earlier age (The solution to modeling both neuritic and neurofibrillary pathology in Alzheimer's disease appears to be to combine mutant APP with mutant tau because tau is the building block of neurofibrillary tangles. Two years ago, several groups showed that mutations in tau protein were associated with frontotemporal dementia with Parkinsonism (FTDP-17) (Clark et al., 1998; Hutton et al., 1998; Poorkaj et al., 1998). These mutations appear to be sufficient to cause disease because mice expressing these mutant tau constructs develop tangles, although the distribution of tangles does not resemble that of FTDP-17 (Lewis et al., 2000). Armed with the mutant tau constructs, Hutton and colleagues showed that mice transgenic for both mutant APP and mutant tau develop neuritic plaques, and neurofibrillary tangles. The Aβ in these animals appeared to accelerate development of the neurofibrillary tangles because the double transgenic mice develop neurofibrillary tangles much more quickly that mice expressing only mutant tau (Lewis et al., 2001). Concurrent studies from Nitsch's group supported this concept because they showed that injected Aβ also accelerated neurofibrillary pathology in mice expressing mutant tau (Gotz et al., 2001). These results emphasize the value of combining transgenes to create murine models that mimic human disease.

    Research in animal models of Parkinson's disease in many ways parallels the development of animal models for Alzheimer's disease. The A53T and A30P mutations in α-synuclein were the first mutations linked to Parkinson's disease (Kruger et al., 1998; Polymeropoulos et al., 1997). Although these mutations cause PD only rarely, further interest in α-synuclein was stimulated by the observation that α-synuclein is one of the major proteins present in Lewy bodies (Spillantini et al., 1997). Based on this observation, several groups generated transgenic mice overexpressing a-synuclein. The results were mixed. A transgenic Drosophila developed by Feany and colleagues appears to model Parkinson's disease closely (Feany and Bender, 2000). This fly shows degeneration of dopaminergic neurons, and develops inclusions in dopaminergic neurons that contain fibrillar α-synuclein resembling Lewy bodies. Transgenic mice over-expressing α-synuclein have been less promising. A transgenic mouse over-expressing α-synuclein that was developed by Masliah, Mucke and colleagues developed dopaminergic dysfunction, showed loss of dopaminergic terminal in the basal ganglia, and showed inclusions containing α-synuclein (Masliah et al., 2000).

    However, this model did not fully recapitulate PD because there was no loss of dopaminergic neurons, as is seen in Parkinson's disease, and the inclusions did not fully resemble Lewy bodies. Transgenic mice over-expressing α-synuclein developed by other groups was less promising. Some lines showed little pathology, and others showed pathology that was not to be selective for dopaminergic neurons, and in none of the cases were structures resembling Lewy bodies observed (Kahle et al., 2000; van der Putten et al., 2000). It is unclear why the phenotype of the mice would differ so greatly among the various lines. Expression level does not correlate with pathology because the mice developed by Masliah and colleagues have less expression of α-synuclein than do other transgenic α-synuclein mice. Genetic background could be important, but this takes time to track down in out-bred mice.

    In the case of Drosophila, a form of genetic background could be important, because Drosophila does not have a homologue of α-synuclein, and so might not have related proteins that can interfere with the aggregation of transgenic human α-synuclein. The reason for this variability in pathology among different transgenic models probably reflects the modest tendency of a-synuclein to aggregation. Although α-synuclein does aggregate, it does not have as strong a tendency to aggregate as Aβ, long stretches of polyglutamine or the cystic fibrosis receptor. Because of its modest tendency to aggregate, aggregation of α-synuclein is highly dependent on environmental conditions.

    A number of factors are known to promote α-synuclein aggregation. Among these is the Aβ peptide. Fragments of α-synuclein are found associated with neuritic plaques in both AD brain and the Tg2576 transgenic mouse overexpressing APP (Ueda et al., 1993; Yang et al., 2000). In vitro studies showed that Aβ promotes α-synuclein aggregation by binding to a domain between amino acids 60-90 of α-synuclein (Paik et al., 1998). Enter the current paper by Masliah and colleagues. They took a transgenic APP mouse, and crossed it with a transgenic α-synuclein mouse (Masliah et al., 2001). The results were striking. The animals developed fibrillar deposits of α-synuclein resembling Lewy bodies throughout the neocortex, and they showed marked degeneration of cholinergic neurons and marked memory loss. This suggests that the presence of large amounts of Aβ promotes aggregation of a-synuclein, in vivo. However, α-synuclein is not known to promote Aβ aggregation, and indeed, the presence of the α-synuclein transgene does not affect the appearance of neuritic plaques. The pathology and cognitive decline of this double transgenic mouse closely resembles the pathology and presentation of DLBD, which is a dementia that shows early and severe memory loss in humans. Thus, the technique of combining two transgenes together in one transgenic mouse has once again yielded a transgenic model that resembles human disease far closer than prior transgenic models.

    Few models are perfect, and this double transgenic also has some puzzling features. The team suggests that the presence of the double transgene "accelerates development of α-synuclein-dependent motor deficits." However, there is no evidence of progressive motor deficits, because the motor deficit present in the double transgenic mice is similar at both six and 12 months. Also, the nature of the fibrillar deposits remains to be explored in more detail. They are clearly visible by immunocytochemistry, but by immunoblotting, the team observes only dimers and trimers of α-synuclein. These problems, however, represent small problems for a transgenic animal model that is otherwise very promising and quite exciting.

    What lies in the future? This research identifies a potentially important link between Aβ and α-synuclein. This link had been proposed earlier by the late Tsunao Saitoh in collaboration with Dr. Masliah because of their discovery of a fragment of α-synuclein in neuritic plaques (Ueda et al., 1993). However, the amount of α-synuclein in neuritic plaques varies greatly from patient to patient, with only about 15% of Alzheimer patients showing α-synuclein in neuritic plaques (Arai et al., 2001). The discovery that mice carrying both APP and α-synuclein transgenes develop a disease resembling DLBD suggests that there is also an important link between Aβ and cortical Lewy body formation. It will be interesting to examine the pathology of DLBD to determine whether there is evidence of Aβ in cortical Lewy bodies in DLBD.

    The success of this work also will stimulate research on further animal models of PD. This focuses on adding two different types of complexity to the existing animal models. One important transgene that is likely to be important is parkin, an E3 ubiquitin ligase whose loss causes delayed degeneration of dopaminergic neurons in neurons (the disease, autosomal juvenile recessive Parkinsonism) (Kitada et al., 1998; Shimura et al., 2000; Zhang et al., 2000). Parkin has been shown to be present in Lewy bodies, bind α-synuclein, and ubiquitinated a glycosylated form of α-synuclein (Choi et al., 2000; Shimura et al., 2001).

    Because of the link between parkin and α-synuclein, many investigators hypothesize that manipulating the expression of parkin in mice overexpressing α-synuclein will affect the tendency of α-synuclein to aggregate, or the ubiquitination of the α-synuclein aggregates that do form. Molecular geneticists are identifying other genes/proteins that play important roles in the susceptibility to PD, and as these genes become identified each gene represents another potential candidate for combining with α-synuclein in a transgenic model.

    The other approach focuses on external factors, such as pesticides, metals and herbicides. PD is known to be associated with agricultural communities. Recently, chronic treatment of rats with the pesticide-mitochondrial toxin, rotenone, was shown to induce formation of structures resembling Lewy bodies (Betarbet et al., 2000). Rotenone has also been shown to induce α-synuclein aggregation in cells overexpressing α-synuclein (Uversky et al., 2001). Combining rotenone with transgenic technologies, for instance by treating transgenic α-synuclein mice with rotenone, might yield even better models of PD, and provide a direct link between the environment and genetics. The information derived from these animal models will ultimately provide profound insights to help us understand what factors cause particular diseases, and what types of treatments might prevent the diseases most effectively.

    Related Links:
    The Biology of α-synuclein and Lewy body disease workshop
    Synuclein and Parkin Roundup from 2000 Society for Neuroscience meeting

     

     

    View all comments by Benjamin Wolozin
  2. A fascinating study which further emphasises the parallels between tau andsynuclein pathobiology: mutations in both genes lead to pathology, haplotypes at both genes predispose to sporadic disease, both pathologies occur 'downstream' in Alzheimer's disease, and now, both pathologies are potentiated in APP mutant mice.