Parkinson's disease is the second most common neurodegenerative disease. It is charaterized by a progressive loss of dopamine (DA) producing neurons in the midbrain, which result in a decline of DA innervations present in the forebrain, in particular, the striatum. The disease leads to appearance of motor symptoms involving akinesia/bradykinesia, gait disturbances, postural imbalance and tremor. Oral administration of L-3,4-dihydroxyphenylalanine (L-DOPA), the precursor of DA, provides very good symptomatic relief, but this intermittent and pharmacological treatment is compromised by severe side effects, such as the appearance of abnormal involuntary movements. Viral vector-mediated direct gene transfer techniques are currently being explored in order to provide continuous and stable synthesis of DA in the brain. This review focuses on the basic idea of DA replacement, first describing the enzymatic machinery important for DA synthesis and secondly the various alternative strategies pursued in several laboratories. The DOPA delivery strategy, based on the co-transduction of tyrosine hydroxylase (TH), and GTP cyclohydrolase 1 (GCH1) genes, has been shown to be a powerful approach providing a robust behavioral recovery and reversal of side effects of the pulsatile administration of L-DOPA medication. The DA delivery strategy, on the other hand, aims at triple transduction of the TH, GCH1 and aromatic amino-acid decarboxylase (AADC) enzymes, and thereby provide a higher rate of conversion of DOPA to DA. Finally, transduction of AADC alone has been proposed as a means to improve the conversion of peripherally administered L-DOPA. As the basic scientific rationale behind these strategies are well understood and the results of the animal experiments are very encouraging, we are now entering into an exciting phase with increasing momentum toward the first clinical applications using this experimental therapy in patients suffering from PD.