Gene Therapy for Dopamine Replacement in Parkinson s Disease

Transcription

1 MEDICINE Gene Therapy for Dopamine Replacement in Parkinson s Disease Anders Björklund, 1 * Tomas Björklund, 2 Deniz Kirik 2 Published 14 October 2009; Volume 1 Issue 2 2ps2 The introduction of L-dopa (L-3,4-dihydroxyphenylalanine) therapy 40 years ago was a revolution in the treatment of patients with Parkinson s disease (PD). With time, however, the shortcomings of oral L-dopa medication became apparent, in particular the appearance of troublesome side effects, expressed as involuntary movements (dyskinesias) that developed over time in many patients. A gene therapy approach, aimed at restoring dopamine synthesis in the affected brain by viral vector delivery of genes that encode the dopamine-synthesizing enzymes, may offer a solution to this problem. Now, a team of French and UK researchers reports promising results in a nonhuman primate model of PD, paving the way for clinical trials of this enzyme-replacement approach. Parkinson s disease (PD) is a multifaceted disorder involving complex neurodegenerative changes that develop over years or decades. The spectrum of symptoms becomes more complex as the disease progresses. At the core of the disease, and central to the cardinal motor symptoms, is the loss of neurons in the substantia nigra region of the midbrain that produce the neurotransmitter dopamine and give rise to the nigrostriatal dopamine system. Dopamine was discovered as a neurotransmitter half a century ago (1 3) and soon after was recognized to be a key regulator of motoric and motivational behavior in the motor-controlling part of the brain: the basal ganglia. Animals deprived of this neurotransmitter are severely impaired in their ability to initiate movements; and in humans suffering from PD, the loss of midbrain dopamine neurons in particular those located in the substantia nigra has a profound impact on the patients ability to move (4, 5). Despite the fact that our understanding of the mechanisms that cause PD has improved considerably in recent years, there is still no cure for this debilitating disorder. In this issue of Science Translational Medicine, Jarraya et al. (6) report the results of a preclinical study on a novel gene therapy approach that aims to restore the dopaminesynthesizing machinery in parkinsonian 1 Wallenberg Neuroscience Center, Department of Experimental Medical Science, Lund University, Lund, Sweden. 2 Brain Repair and Imaging in Neural Systems (B.R.A.I.N.S), Department of Experimental Medical Science, Lund University, Lund, Sweden. *Corresponding author. , anders.bjorklund@ med.lu.se monkeys. In this Perspective, we review the rationale underlying this proposed restorative therapy in the context of other, alternative enzyme-replacement approaches that are now being developed (7). Dopamine is synthesized from the essential amino acid l-tyrosine in two steps, first by hydroxylation to l-dopa (l-3,4-dihydroxyphenylalanine) by the tyrosine hydroxylase (TH) enzyme, and then by the conversion of l-dopa to dopamine by aromatic l amino acid decarboxylase (AADC). In line with the pioneering work of Carlsson and Hornykiewicz (1, 3, 8), the neurochemical changes underlying the core motor symptoms of PD can be viewed as a breakdown of the enzymedependent machinery that synthesizes and releases dopamine (Fig. 1). This view is supported by the finding that systemic administration of l-dopa is effective in restoring dopamine production in the affected brain region, the striatum, and that this restoration is accompanied by substantial symptomatic relief in patients with manifest disease. Dopamine replacement therapy, accomplished by the treatment of patients with l-dopa, was introduced into the clinic in the late 1960s (9) and continues to be the mainstay of medical therapy for PD. However, chronic treatment of patients with orally delivered l-dopa can give rise to serious side effects, expressed as involuntary movements, dyskinesias, that develop over time. As a consequence, the difference between the l-dopa dose that yields a therapeutic benefit and the one that induces dyskinesia the so-called therapeutic window may be lost completely (10 12). There is considerable evidence that the dyskinetic side effects of l-dopa medication are caused by the intermittent, pulsatile oral delivery and the associated excessive swings in extracellular dopamine concentrations. Approaches to continuous delivery of l-dopa, by intragastric, duodenal, or intravenous infusion, have indeed proved efficient in ameliorating the dyskinetic side effects (12 14). These insights have inspired the concept of using viral vector delivery of the genes that encode TH and AADC into the brain (a process referred to as gene transfer) as a tool for obtaining a local source of continuous and more physiological dopamine delivery to the striatum. This approach should help patients avoid the side effects associated with standard oral l-dopa medication (15, 16). Two versions of this gene therapy approach have now reached the clinic. In an ongoing phase I/II clinical trial, the UKbased company Oxford Biomedica is testing a single lentiviral vector (LV) that contains the genes that encode all three enzymes needed for efficient dopamine synthesis: the TH and AADC enzymes and the tetrahydrobiopterin (BH4) cofactor producing enzyme GTP-cyclohydrolase-1 (GCH1) (Fig. 2B) (17). In the article published in this issue of Science Translational Medicine, a team of French scientists, working in collaboration with Oxford Biomedica researchers, have presented the key preclinical data underlying the Oxford Biomedica trial. These data were obtained from a study performed in nonhuman primates in which a PD-like syndrome was induced by systemic injections of the dopamine neurotoxin MPTP (1-methyl- 4-phenyl-1,2,3,6-tetrahydropyridine) (6). In an alternative approach, a research group at the University of California, San Francisco (UCSF), sponsored by the U.S.-based company Genzyme, has performed a small clinical trial in five PD patients using an adenoassociated virus (AAV) vector that encodes only one of the three enzymes, AADC (Fig. 2A) (7). The preclinical studies that led to this trial are available elsewhere (18 20). What are the relative virtues of these two alternative approaches? The choice of the UCSF/Genzyme team to focus on AADC alone is based on the consideration that the efficiency of oral l-dopa therapy is critically dependent on the capacity of the brain to convert l-dopa into dopamine, the reaction catalyzed by AADC. Normally this conversion takes place in the dopaminergic axonal terminals, where the bulk of the AADC enzyme is localized (Fig. 1). In postmortem analyses of brains from PD 14 October 2009 Vol 1 Issue 2 2ps2 1

2 Fig. 1. Dopamine neurotransmission in the normal brain. Under normal physiological conditions, tyrosine-derived dopamine is synthesized, stored in, and released from the axon terminals in the striatum (shown) derived from the nigrostriatal dopaminergic neurons. Red dots, dopamine; DAT, dopamine transporter. patients, however, the AADC enzyme concentration is greatly reduced relative to that in brains from control individuals (21, 22). These findings suggest that in advanced cases of PD, the remaining decarboxylating capacity may no longer be sufficient to sustain the therapeutic efficacy of oral l-dopa medication. Thus, the strategy underlying the UCSF/Genzyme trial is to restore the capacity for l-dopa conversion in patients in which this treatment has begun to fail (Fig. 2A). The attraction of this pro-drug gene transfer approach is that the expressed enzyme should be functional only when l-dopa is administered, which provides a level of safety and also the advantage that the functional effect could be regulated by adjustment of the oral l-dopa dose. The initial data obtained in the five patients that were given a low dose of the AAV-AADC vector (injected bilaterally in the posterior putamen region of the brain) support safety and showed a sustained 30% increase in striatal decarboxylating activity (as determined by positron emission tomography) (7). Clinically relevant therapeutic effects, however, may require higher doses and more widespread distribution of the vector. The goal of the Oxford Biomedica approach explored in the Jarraya et al. study (6) is to establish local dopamine production in the brain through expression of the dopamine-biosynthetic enzymes TH, AADC, and GCH1 (Fig. 2B). The genes encoding these enzymes were introduced as a single tri-cistronic construct using an equine infectious anemia virus (EIAV) vector (a LV vector of nonhuman origin) (23). Inclusion of the GCH1 gene in this triple combination is necessary, because the TH enzyme requires the BH4 cofactor in order to catalyze the conversion of l-tyrosine to dopa (24). The endogenous concentrations of BH4 are much reduced in the dopaminedenervated striatum, and as a consequence, the availability of this cofactor seriously limits the efficacy of dopa synthesis by ectopically expressed TH (25). In addition, in order to avoid shutdown of the TH enzyme by the negative feedback inhibition exerted by free cytosolic dopamine (26, 27), the TH construct used here is a truncated version that lacks the regulatory N-terminal portion of the protein. Ozawa and co-workers (28, 29) were the first to explore this triple-enzyme delivery approach using of a mixture of three AAV vectors, each encoding one of the three enzymes. Their results show that it is indeed possible to obtain sustained functional levels of putaminal dopamine production in rodent and nonhuman primate models of PD. In the Jarraya et al. study (6), this feat has now been achieved with a single EIAV vector construct that was injected bilaterally into the putamen. The results show a substantial improvement in mobility and a reduction in disability scores that manifested gradually over the first 6 weeks after vector injection and were sustained for a long period of time, more than a year in some of the treated animals. The dyskinesia induced by daily oral l-dopa intake was also reduced in these animals. These results are encouraging and provide important proof of concept for the multiple enzyme replacement strategy in the treatment of PD. Restoration of the complete dopamine-synthetic machinery by the tripleenzyme approach may seem logical, but it has some potentially important caveats. The EIAV or AAV vectors, when injected into the striatum, transduce almost exclusively the intrinsic striatal neurons. Because these cells use GABA (γ-aminobutyric acid) as their neurotransmitter, they do not normally produce dopamine (Fig. 2B). Therefore, they lack the appropriate vesicular storage and release mechanisms that are normally present in dopamine neurons. Nonvesicular, cytosolic dopamine is not only ineffectively released from these cells but also may cause problems. Cytosolic dopamine can act as a feedback inhibitor of the TH enzyme and a competitive inhibitor of the binding of the cofactor BH4 to the active site of TH (27). Although the truncated version of TH used in the EIAV vector is constitutively active, the end-product inhibition exerted by cytosolic dopamine will remain, at least in part. Furthermore, it is conceivable that the accumulation of cytosolic dopamine will expose the transduced cells to excessive oxidative stress, possibly leading to toxic damage or neurodegeneration (30). In addition, data obtained by Bankiewicz and co-workers in MPTP-treated monkeys have raised the concern that nonregulated dopamine production induced by the overexpression of AADC may trigger dyskinesias (18). The low endogenous level of this enzyme present in the parkinsonian striatum is rate-limiting for the production of dopamine from exogenous l-dopa. Although this situation may restrict the functional efficacy of dopa therapy, it 14 October 2009 Vol 1 Issue 2 2ps2 2

3 Fig. 2. Two modes of gene therapy for dopamine replacement in PD. These two modalities of enzyme-encoding gene delivery explored for local dopamine production or dopa delivery in PD are based on a similar idea, but they are conceptually and mechanistically quite different. (A) The single-enzyme pro-drug approach. In this mode, explored in the UCSF/Genzyme trial (7), only the decarboxylating enzyme, AADC, is expressed from the vector, and dopamine is formed in host striatal neurons from L-dopa supplied by the normal oral route. With this approach, the enzyme is expected to be functional only when L-dopa is administered, thus providing a possibility for regulation of the effect by adjustment of the L-dopa dose. Red dots, dopamine. (B) The triple-enzyme approach. The first mode pursued in the Jarraya et al. study (6) (and the now-ongoing Oxford Biomedica trial) is to supply all three enzymes needed for dopamine synthesis, TH, AADC, and GCH1, to the dopamine-deficient host striatum. With this approach, the three enzymes are expressed within the same cells, the resident striatal GABAergic neurons, which allows the enzymes to synthesize DOPA from tyrosine and convert it to dopamine in a continuous manner. However, cytoplasmic dopamine generated within neurons that do not possess the normal machinery for dopamine storage and release is likely to be metabolized quickly and may thus be rather inefficient. also provides a safety mechanism that keeps the dopamine production rate in check. These problems are avoided by removing the AADC gene from the vector construct. In such a two-enzyme approach, performing gene transfer of the TH and GCH1 genes only, the end product is dopa, and the synthesis of dopamine relies on the endogenous AADC activity in the host striatum. The argument raised against this approach is that the residual AADC activity may be too low to generate functional amounts of dopamine in the denervated striatum, at least in some of the most advanced cases of PD. Studies of local dopa delivery in rodent models of PD by AAV vector mediated transfer of the TH and GCH1 genes into the striatum suggest that complete reversal of the PD-associated motor deficits may be achieved more readily if the endogenous dopamine innervation is partially spared, but that substantial symptomatic improvement also can be obtained in animals with complete nigrostriatal dopamine denervation (31 33). In the dopamine-denervated striatum, the functional efficacy of vector-mediated dopa delivery takes advantage of the capacity of the serotonin innervation, which is at least partially spared in PD, to convert dopa to dopamine and to store and release the newly synthesized dopamine in an impulse-dependent manner (Fig. 3). With intermittent oral l-dopa medication, the serotonin neurons may play an important role in induction of the dyskinesia that results from uncontrolled, excessive swings in dopa-derived dopamine, released as a false transmitter from serotonin terminals (34). This should not be a problem when dopa is supplied continuously at low, constant concentrations into the extracellular space from the transduced cells in the striatum. In such cases, dopamine released 14 October 2009 Vol 1 Issue 2 2ps2 3

4 from serotonin neurons at stable physiological concentrations should be beneficial rather than detrimental. In addition, the physical separation of the site of dopa synthesis (in the transduced striatal neurons) from the site of dopamine synthesis and release (in spared dopamine and serotonin axon terminals) should be advantageous in that it avoids the potential complications associated with nonregulated dopamine production within the transduced striatal neurons. The clinical trials now under way reflect the outcome of extensive experimental work carried out in rodent and primate models of PD over the past 15 years. The concept is attractive: Restoration of the dopaminesynthetic machinery in a brain region that has lost this capacity is a logical, and technically refined, extension of standard l-dopa pharmacotherapy. The proven efficacy and safety of oral l-dopa medication, and the problems caused by intermittent pulsatile l-dopa delivery over the long term, provide a compelling rationale for the continued development of the enzyme-replacement approach. Initially, the main hurdle was to achieve sufficient levels of dopa or dopamine production in the absence of vector-induced toxicity. During the early years of the development of recombinant viruses for gene delivery, the vectors were simply not good enough to fulfill these requirements. Thanks to further refinements of the constructs and vector production methods, however, a new generation of efficient and essentially nontoxic AAV and LV vectors, suitable for gene delivery to the nervous system, has been developed. The animal experimental work shows that these delivery systems hold great promise, not only as experimental tools but also for use in patients. The Oxford Biomedica trial (17) is a first in three respects: It is the first to test the triple-enzyme approach in PD, the first to test a LV-based retroviral vector in the Fig. 3. The two-enzyme approach to dopa replacement gene therapy for PD. A third enzyme-encoding gene-delivery approach for local dopamine production is based on ideas similar to those in Fig. 2 (A and B), but again is conceptually and mechanistically quite different. A potential drawback of both approaches shown in Fig. 2 (A and B), is that cytosolic dopamine, generated in cells that lack vesicular storage, may expose the transduced cells to excessive oxidative stress that may lead to toxic damage. This problem is avoided in the two-enzyme approach, in which only the dopa-producing enzymes TH and GCH1 are expressed. In this case, the synthesis of dopamine will take place in the host striatum, mostly in remaining dopamine synapses and as a false transmitter in the spared serotonin terminals. This has the advantage that the dopa-derived dopamine can be stored and released in a normal, physiological, and impulse-dependent manner, similar to the situation in the intact striatum (shown in Fig. 1). This mode of delivery, however, is critically dependent on the presence of sufficient amounts of endogenous AADC enzyme activity in the host striatum. This may not be a problem in less-affected patients, but could limit the efficacy of this approach in more advanced stages of PD, particularly in cases in which both the dopaminergic and serotonergic innervations have been lost completely. Red dots, Dopamine; Blue dots, Serotonin. human brain, and the first to test a viral vector of nonhuman origin in clinical gene therapy. The pioneering nature of this trial is not without risks, and it will no doubt be carefully watched by the wider gene therapy community. Side effects seen in clinical trials using integrating retroviral vectors for applications outside the central nervous system have raised safety concerns that also must be taken seriously in neurological applications (35). However, as with other biologics, long-term efficacy and safety can only be fully assessed in carefully monitored trials in patients. The translation of gene therapy approaches from animal models to human disease involves a number of key issues related to vector dosing, up-scaling of vector delivery to accommodate the large size of the human brain, immunotoxicity, and safety. These issues have to be addressed cautiously, step by step, in order to avoid serious adverse 14 October 2009 Vol 1 Issue 2 2ps2 4

5 events that may jeopardize the development of effective gene therapies in the neurological field. REFERENCES 1. A. Carlsson, M. Lindqvist, T. Magnusson, 3,4-Dihydroxyphenylalanine and 5-hydroxytryptophan as reserpine antagonists. Nature 180, 1200 (1957). 2. A. Carlsson, M. Lindqvist, T. Magnusson, B. Waldeck, On the presence of 3-hydroxytyramine in brain. Science 127, 471 (1958). 3. O. Hornykiewicz, Dopamine (3-hydroxytyramine) in the central nervous system and its relation to the Parkinson syndrome in man (translation from German). Deutsch. Med. Wochenschr. 87, (1962). 4. S. B. Dunnett, Motor function(s) of the nigrostriatal dopamine system: Studies of lesions and behavior, in Handbook of Chemical Neuroanatomy Vol. 21, Dopamine, S. B. Dunnett, M. Bentivoglio, A. Björklund, T. Hökfelt, Eds. (Elsevier, Oxford, 2005), pp J. W. Langston, MPTP: Insights into the etiology of Parkinson s disease. Eur. Neurol. 26 (suppl. 1), 2 10 (1987). 6. S. Jarraya, S. Boulet, G. S. Ralph, C. Jan, G. Bonvento, M. Azzouz, J. E. Miskin, M. Shin, T. Delzescaux, X. Drouot, A.-S. Hérard, D. M. Day, E. Brouillet, S. M. Kingsman, P. Hantraye, K. A. Mitrophanous, N. D. Mazarakis, S. Palfi, Dopamine gene therapy for Parkinson s disease in a nonhuman primate without associated dyskinesia. Sci. Transl. Med. 2, 2ra4 (2009). 7. J. L. Eberling, W. J. Jagust, C. W. Christine, P. Starr, P. Larson, K. S. Bankiewicz, M. J. Aminoff, Results from a phase I safety trial of haadc gene therapy for Parkinson disease. Neurology 70, (2008). 8. J. K. Andersen, Arvid Carlsson: An early pioneer in translational medicine. Sci. Transl. Med. 1, 2ps3 (2009). 9. G. C. Cotzias, M. H. Van Woert, L. M. Schiffer, Aromatic amino acids and modification of parkinsonism. N. Engl. J. Med. 276, (1967). 10. M. M. Mouradian, J. L. Juncos, G. Fabbrini, T. N. Chase, Motor fluctuations in Parkinson s disease: Pathogenetic and therapeutic studies. Ann. Neurol. 22, (1987). 11. J. A. Obeso, C. W. Olanow, J. G. Nutt, Levodopa motor complications in Parkinson s disease. Trends Neurosci. 23 (suppl.), S2 S7 (2000). 12. C. W. Olanow, J. A. Obeso, F. Stocchi, Continuous dopamine-receptor treatment of Parkinson s disease: Scientific rationale and clinical implications. Lancet Neurol. 5, (2006). 13. D. Nyholm, H. Askmark, C. Gomes-Trolin, T. Knutson, H. Lennernäs, C. Nyström, S.-M. Aquilonius, Optimizing levodopa pharmacokinetics: Intestinal infusion versus oral sustained-release tablets. Clin. Neuropharmacol. 26, (2003). 14. M. M. Mouradian, I. J. Heuser, F. Baronti, T. N. Chase, Modification of central dopaminergic mechanisms by continuous levodopa therapy for advanced Parkinson s disease. Ann. Neurol. 27, (1990). 15. T. Björklund, D. Kirik, Scientific rationale for the development of gene therapy strategies for Parkinson s disease. Biochim. Biophys. Acta 1792, (2009). 16. M. M. Mouradian, T. N. Chase, Gene therapy for Parkinson s disease: An approach to the prevention or palliation of levodopa-associated motor complications. Exp. Neurol. 144, (1997). 17. Oxford Biomedica announces update on phase I/II study of ProSavin in Parkinson s disease, 13 July 2009 ( newsid=236). 18. K. S. Bankiewicz, M. Daadi, P. Pivirotto, J. Bringas, L. Sanftner, J. Cunningham, J. R. Forsayeth, J. L. Eberling, Focal striatal dopamine may potentiate dyskinesias in parkinsonian monkeys. Exp. Neurol. 197, (2006). 19. K. S. Bankiewicz, J. Forsayeth, J. L. Eberling, R. Sanchez- Pernaute, P. Pivirotto, J. Bringas, P. Herscovitch, R. E. Carson, W. Eckelman, B. Reutter, J. Cunningham, Long-term clinical improvement in MPTP-lesioned primates after gene therapy with AAV-hAADC. Mol. Ther. 14, (2006). 20. J. R. Forsayeth, J. L. Eberling, L. M. Sanftner, Z. Zhen, P. Pivirotto, J. Bringas, J. Cunningham, K. S. Bankiewicz, A dose-ranging study of AAV-hAADC therapy in Parkinsonian monkeys. Mol. Ther. 14, (2006). 21. T. Nagatsu, T. Yamamoto, T. Kato, A new and highly sensitive voltammetric assay for aromatic l-amino acid decarboxylase activity by high-performance liquid chromatography. Anal. Biochem. 100, (1979). 22. K. Lloyd, O. Hornykiewicz, Parkinson s disease: Activity of l-dopa decarboxylase in discrete brain regions. Science 170, (1970). 23. M. Azzouz, E. Martin-Rendon, R. D. Barber, K. A. Mitrophanous, E. E. Carter, J. B. Rohll, S. M. Kingsman, A. J. Kingsman, N. D. Mazarakis, Multicistronic lentiviral vector-mediated striatal gene transfer of aromatic l-amino acid decarboxylase, tyrosine hydroxylase, and GTP cyclohydrolase I induces sustained transgene expression, dopamine production, and functional improvement in a rat model of Parkinson s disease. J. Neurosci. 22, (2002). 24. C. Bencsics, S. R. Wachtel, S. Milstien, K. Hatakeyama, J. B. Becker, U. J. Kang, Double transduction with GTP cyclohydrolase I and tyrosine hydroxylase is necessary for spontaneous synthesis of l-dopa by primary fibroblasts. J. Neurosci. 16, (1996). 25. O. Corti, A. Sánchez-Capelo, P. Colin, N. Hanoun, M. Hamon, J. Mallet, Long-term doxycycline-controlled expression of human tyrosine hydroxylase after direct adenovirus-mediated gene transfer to a rat model of Parkinson s disease. Proc. Natl. Acad. Sci. U.S.A. 96, (1999). 26. S. R. Wachtel, C. Bencsics, U. J. Kang, Role of aromatic l- amino acid decarboxylase for dopamine replacement by genetically modified fibroblasts in a rat model of Parkinson s disease. J. Neurochem. 69, (1997). 27. S. C. Kumer, K. E. Vrana, Intricate regulation of tyrosine hydroxylase activity and gene expression. J. Neurochem. 67, (1996). 28. S.-I. Muramatsu, K.-I. Fujimoto, K. Ikeguchi, N. Shizuma, K. Kawasaki, F. Ono, Y. Shen, L. Wang, H. Mizukami, A. Kume, M. Matsumura, I. Nagatsu, F. Urano, H. Ichinose, T. Nagatsu, K. Terao, I. Nakano, K. Ozawa, Behavioral recovery in a primate model of Parkinson s disease by triple transduction of striatal cells with adeno-associated viral vectors expressing dopamine-synthesizing enzymes. Hum. Gene Ther. 13, (2002). 29. Y. Shen, S. I. Muramatsu, K. Ikeguchi, K. I. Fujimoto, D. S. Fan, M. Ogawa, H. Mizukami, M. Urabe, A. Kume, I. Nagatsu, F. Urano, T. Suzuki, H. Ichinose, T. Nagatsu, J. Monahan, I. Nakano, K. Ozawa, Triple transduction with adeno-associated virus vectors expressing tyrosine hydroxylase, aromatic-l-amino-acid decarboxylase, and GTP cyclohydrolase I for gene therapy of Parkinson s disease. Hum. Gene Ther. 11, (2000). 30. L. Chen, Y. Ding, B. Cagniard, A. D. Van Laar, A. Mortimer, W. Chi, T. G. Hastings, U. J. Kang, X. Zhuang, Unregulated cytosolic dopamine causes neurodegeneration associated with oxidative stress in mice. J. Neurosci. 28, (2008). 31. D. Kirik, B. Georgievska, C. Burger, C. Winkler, N. Muzyczka, R. J. Mandel, A. Björklund, Reversal of motor impairments in parkinsonian rats by continuous intrastriatal delivery of l-dopa using raav-mediated gene transfer. Proc. Natl. Acad. Sci. U.S.A. 99, (2002). 32. T. Carlsson, C. Winkler, C. Burger, N. Muzyczka, R. J. Mandel, A. Cenci, A. Björklund, D. Kirik, Reversal of dyskinesias in an animal model of Parkinson s disease by continuous l-dopa delivery using raav vectors. Brain 128, (2005). 33. L. Leriche, T. Björklund, N. Breysse, L. Besret, M. C. Grégoire, T. Carlsson, F. Dollé, R. J. Mandel, N. Déglon, P. Hantraye, D. Kirik, Positron emission tomography imaging demonstrates correlation between behavioral recovery and correction of dopamine neurotransmission after gene therapy. J. Neurosci. 29, (2009). 34. M. T. Carta, T. Carlsson, D. Kirik, A. Björklund, Dopamine released from 5-HT terminals is the cause of l-dopainduced dyskinesia in parkinsonian rats. Brain 130, (2007). 35. D. A. Williams, Gene therapy continues to mature and to face challenges. Mol. Ther. 17, (2009) /scitranslmed Citation: A. Björklund, T. Björklund, D. Kirik, Gene therapy for dopamine replacement in Parkinson s disease Sci. Transl. Med. 1, 2ps2 (2009) October 2009 Vol 1 Issue 2 2ps2 5