news and views

temperature, and it was only the development of laser cooling techniques during the past two decades, allowing atoms to be cooled to microkelvin temperatures, that made optical trapping of atoms possible. These achievements were honoured with the award of the 1997 Nobel Prize in Physics to Chu, CohenTannoudji and Phillips. Quantum physics teaches us that the light field is quantized; that is, light comes in energy units of E = h, where h is Plancks constant and is the light frequency. These light quanta are called photons. The granular quantum structure of light becomes visible only when we deal with extremely weak light fields consisting of a few photons. In this sense, a single photon represents the weakest possible light field. So how can the light intensity of a single photon be sufficient to trap an atom? The key idea is to store the photon in an optical cavity consisting of two high-quality mirrors. The smaller the volume in which we confine the photon, the larger is the energy per volume, and it is the corresponding strong electric field that is seen by the atom and determines the optical force. Trapping an atom with a single photon can thus be achieved using microcavities with extremely high-quality optical mirrors, where the distance between the mirrors is only ten or twenty times the optical wavelength. This is the basis of the Max-Planck and Caltech experiments13. How are atoms actually captured inside the cavity, and how do we know that it is there? The cavity mirrors are not perfect, and a single photon will be lost from the cavity after a short time. These photons must therefore be continuously replenished by a pumping laser that puts the photon back into the cavity. But a unique feature of such cavity QED experiments is that the light field leaking out of the cavity allows us to observe the motion of the atom in real time inside the cavity. So we can see an atom entering the cavity, and this enables a feedback loop to change the parameters guiding the laser, which can then capture the atoms. The latest developments in cavity QED with optical and microwave photons should be seen in a much broader context. An important new direction in modern physics has been the attempt to control processes in individual quantum systems on a quantum by quantum level. One of the long-term hopes of these endeavours is to develop the ability to assemble quantum devices as composite controllable quantum systems. Quantum optics has taken a leading role in this area, and cavity QED and laser-cooled trapped ions are its prime examples. The recent achievements in cavity QED15 and ion traps6 now set the stage for future developments, ranging from basic studies of quantum physics all the way to possible applications. Among the most exciting prospects are
NATURE | VOL 404 | 23 MARCH 2000 | www.nature.com

the possibilities offered by cavity QED to build quantum information processors and to connect these devices in an optical quantum network7. Quantum computers require the storage of information in a set of two-level systems (called quantum bits or qubits), the processing of this information using quantum gates, and a final readout by a quantum-state measurement. In a typical cavity QED set up, long-lived internal states of atoms can play the role of reliable quantum memory for storing the qubits. Quantum operations on a single qubit can be performed by coupling two atomic states with laser pulses. Joint operations on two qubits require the controlled exchange of a cavity photon between two atoms, where the cavity mode plays the role of a quantum data bus. This mechanism for a two-bit gate relies directly on the strong coupling of single photons with single atoms. Although isolated atoms represent an ideal quantum memory, optical photons propagating in fibres are a fast and reliable
Neurodegenerative diseases

way to transmit quantum information to distant nodes of a network (Fig. 1). Cavity QED provides a natural setting for the interface between atoms and photons in the system in the form of optical interconnects. The primary technical obstacle to achieving these goals is the need to develop techniques to confine atoms within cavities. The recent experiments by the Caltech and Max-Planck groups are important milestones in this direction. s
Peter Zoller is at the Institute for Theoretical Physics, University of Innsbruck, 6020 Innsbruck, Austria. e-mail: [email protected]
1. Pinkse, P. W. H., Fischer, T., Maunz, P. & Rempe, G. Nature 404, 365368 (2000). 2. Hood, C. J., Lynn, T. W., Doherty, A. C., Parkins, A. S. & Kimble, H. J. Science 287, 14471453 (2000). 3. Ye, J., Vernooy, D. W. & Kimble, H. J. Phys. Rev Lett. 83, 49874990 (1999). 4. Nogues, G. et al. Nature 400, 239242 (1999). 5. Varcoe, B. T. H., Brattke, S., Wedinger, M. & Walther, H. Nature 402, 743746 (2000). 6. Sackett, C. et al. Nature 404, 256259 (2000). 7. Cirac, J. I., Zoller, P., Kimble, H. J. & Mabuchi, H. Phys. Rev. Lett. 78, 32213224 (1997).

Parkinsons pathology in a fly

Christian Haass and Philipp J. Kahle
uman life expectancy has increased dramatically over the past century. But can we really enjoy our extra years? Diseases such as Alzheimers and Parkinsons, characterized by the death of certain nerve-cell populations and consequent dementia or movement defects, threaten more and more elderly people in the Western world. Many of the molecules behind these diseases have been identified, but we still lack a coherent understanding of the processes involved. On page 394 of this issue1, however, Feany and Bender offer a new way to tackle the problem. They have developed a model of Parkinsons disease in the fruitfly Drosophila melanogaster that reproduces many of the features of the human disorder. The hope is that, with this model, we can learn much about the disease and, perhaps, new ways to treat it. Parkinsons disease is the second most common neurodegenerative disorder. The pathological hallmark of the disease is the accumulation of fibrous protein deposits in neuronal cytoplasm (Lewy bodies) and nerve fibres (Lewy neurites) in the brain. These deposits may interfere with normal neuronal function. Selective death of the neurons that normally secrete the neurotransmitter dopamine results in a movement disorder that is characterized by muscle rigidity and resting tremor. The neurons that are affected in patients with Parkinsons disease are found in the regions of the brain

called the substantia nigra and the locus coeruleus (and form the nigrostriatal dopamine system). Mice, flies and worms that have been engineered to express human genes linked to Alzheimers disease have already yielded invaluable insight into the molecular mechanisms underlying this disorder2, and have even offered exciting possibilities for therapy3. Until recently, it was difficult to take the same tack for Parkinsons disease, as genes linked to its onset or progression were unknown. Three years ago, however, by studying an ItalianGreek family in which Parkinsons disease was inherited, Polymeropoulos et al.4 identified a critical mutation in the gene encoding -synuclein. This mutation resulted in alanine being replaced by threonine at amino-acid position 53 of the protein (A53T mutation). A second mutation, resulting in the substitution of alanine with phenylalanine at position 30 (A30P mutation), was later identified in a German family by Krger et al.5 (Fig. 1a, overleaf). Interestingly, -synuclein, a phosphorylated protein6 normally found in presynaptic neurons7, is the main component of Lewy bodies8 (Fig. 1b, overleaf). Ablation of the gene encoding -synuclein in mice results in functional deficits of the nigrostriatal dopamine system9, providing further evidence for a connection of -synuclein with Parkinsons disease. In familial Alzheimers disease, mutations in several distinct genes
341

2000 Macmillan Magazines Ltd

news and views

accelerate the aggregation and deposition of the amyloid- peptide10, the pathological hallmark of this disease. Likewise, it was expected that the -synuclein mutations might enhance the formation of Lewy bodies, and indeed, early in vitro experiments showed that -synuclein could selfaggregate and form fibres11,12. These findings provided the basis for the generation of an animal model that recapitulates the hallmarks of human Parkinsons disease, a task now achieved using Drosophila by Feany and Bender1. The authors expressed the gene encoding human synuclein (as a transgene) in all Drosophila nerve cells. This led to an age-dependent loss of dopamine-secreting (dopaminergic) neurons; other neurons appeared to be largely unaffected. Death of dopaminergic cells was observed upon expression of wildtype -synuclein or of either of the two mutants (A53T or A30P) identified previously. Strikingly, Feany and Bender also found that some neurons accumulated intracellular aggregates that closely resembled the Lewy bodies found in Parkinsons disease patients. Most important, the inclusions were composed of -synuclein filaments 7 to 10 nanometres in diameter again, characteristics of human Lewy bodies. So, the transgenic flies showed many of the physical traits of patients with Parkinsons disease. But did the flies also show the behavioural characteristics seen in humans? To address this question, Feany and Bender used a climbing test an assay originally designed to follow age-related changes in Drosophila. When tapped down to the bottom of a vial, flies rapidly climb up to the top in what is known as a negative geotactic response. Feany and Bender found that young flies overexpressing -synuclein performed well in this test, but aged transgenic flies frequently fell back to the bottom. Flies expressing wild-type -synuclein or the A53T mutant performed similarly to each other in this test, but locomotor defects were more severe in flies expressing the A30P mutant. This may indicate that the A30P mutant is more aggressive than the A53T mutant, although the possibility that there are small differences in transgene expression cannot be excluded. Nevertheless, this fly model clearly replicates many of the characteristic features of Parkinsons disease. Many laboratories are also trying to generate a transgenic mouse model of Parkinsons disease. Masliah et al.13 have now developed transgenic mice that express wildtype -synuclein under the control of the promoter of the platelet-derived growth factor gene, which is expressed in all neurons. These mice accumulate inclusions composed of -synuclein, although the inclusions do not contain fibres. The amorphous inclusions are seen not only in neuronal cell bodies but also in nuclei of transgenic mice, where they do not occur in Parkinsons disease. Some neurons appear to react with antibodies directed towards a protein called ubiquitin a feature of many classical Lewy bodies in the brains of Parkinsons patients. The transgenic animals also show dopaminergic deficits. Amounts of tyrosine hydroxylase (an enzyme required for the biosynthesis of dopamine) are significantly reduced in animals with excessive -synuclein expression. Moreover, behavioural tests revealed locomotor impairments in these mice. In other mouse models, pan-neuronal expression of the A30P -synuclein mutant led to pathological defects in neuronal extensions (neurites)14,15. Classical Lewy bodies have not yet been observed in these models. But studies of these mice have revealed something interesting about the A30P mutant. Wild-type -synuclein has been reported to bind to cell membranes in vitro, whereas mutant -synuclein does not16. In these mouse models, however, the A30P mutant does not seem to lose this ability entirely, so it shows a toxic gain, rather than loss, of function15. Now that models for Parkinsons disease have been developed in vertebrate and invertebrate animals, a broad range of investigative options has been opened up. Researchers will be able to study the influence of pathological protein expression on neuronal function and animal behaviour. A fundamental question can be addressed: whether or not deposition of -synuclein causes selective neuronal death in Parkinsons disease. And finally, it will now be possible to screen for compounds that might interfere with pathological events in Parkinsons disease. In this respect, the short generation time of flies makes them invaluable tools for drug screening, and our in-depth knowledge of their genetics will aid in the identification of genes that affect the onset and progression of this debilitating disease. s
Christian Haass and Philipp J. Kahle are at the Adolf-Butenandt-Institute, Department of Biochemistry, Ludwig-Maximilians-University Munich, Schillerstrasse 44, 80336 Munich, Germany. e-mails: [email protected] [email protected]
1. Feany, M. B. & Bender, W. W. Nature 404, 394398 (2000). 2. Price, D. L., Sisodia, S. S. & Borchelt, D. R. Science 282, 10791083 (1998). 3. Schenk, D. et al. Nature 400, 173177 (1999). 4. Polymeropoulos, M. H. et al. Science 276, 20452047 (1997). 5. Krger, R. et al. Nature Genet. 18, 106108 (1998). 6. Okochi, M. et al. J. Biol. Chem. 275, 390397 (2000). 7. Maroteaux, L. & Scheller, R. H. Brain Res. Mol. Brain Res. 11, 335343 (1991). 8. Spillantini, M. G. et al. Nature 388, 839840 (1997). 9. Abeliovich, A. et al. Neuron 25, 239252 (2000). 10. Selkoe, D. J. Nature 399, A23A31 (1999). 11. Conway, K. A., Harper, J. D. & Lansbury, P. T. Nature Med. 4, 13181320 (1998). 12. Narhi, L. et al. J. Biol. Chem. 274, 98439846 (1999). 13. Masliah, E. et al. Science 287, 12651269 (2000). 14. Hsiao, K. Ann. NY Acad. Sci. (in the press). 15. Kahle, P. J. et al. Ann. NY Acad. Sci. (in the press). 16. Jensen, P. H. et al. M. J. Biol. Chem. 273, 2629226294 (1998). 17. Ueda, K. et al. Proc. Natl Acad. Sci. USA 90, 1128211286 (1993).

a
-Synuclein

Repeat region 1 2 3 4 5 6

NAC

Charged carboxy-terminal domain

A30P

A53T

Lewy neurite

Lewy body

-Synuclein -Synuclein

Figure 1 A central role for -synuclein in Parkinsons disease. a, The structure of the -synuclein protein. The potentially lipid-binding amino-terminal domain consists of an imperfectly repeated motif, represented by the numbers 16. The NAC domain corresponds to the non-amyloid component originally isolated from senile plaques of patients with Alzheimers disease17. The two mutations associated with familial Parkinsons disease are indicated. In the A30P mutant, alanine at amino-acid position 30 is mutated to proline; in the A53T mutant, alanine at position 53 is replaced by threonine. b, Lewy neurites (left) and Lewy bodies (right) composed of -synuclein are the characteristic lesions observed in the brains of patients with Parkinsons disease. Similar inclusions of -synuclein are observed in Feany and Benders fly model of Parkinsons disease1. These sections were stained with a polyclonal antibody to human -synuclein6.
NATURE | VOL 404 | 23 MARCH 2000 | www.nature.com

2000 Macmillan Magazines Ltd

343