The neurotransmitter dopamine is critical in the pathophysiology of Parkinson's disease (PD). The seminal discovery that reductions in dopamine produce the cardinal motor symptoms of PD led to a breakthrough treatment: the dopamine precursor L-DOPA. But more breakthroughs are needed. PD is now widely recognized as a multifaceted disorder, including motor and non-motor symptoms. While many of the core motor symptoms of PD respond well to L-DOPA, non-motor symptoms, such as cognitive impairment, are more challenging to treat. Dopamine replacement therapy does not resolve all PD symptoms, and chronic treatment often causes complications such as levodopa-induced dyskinesia. These shortcomings may be related to the importance of dopamine dynamics across space and time. L-DOPA and other pharmacological therapies cannot replicate the precisely calibrated targeting and timing of dopamine release supporting healthy behavior and dopamine-dependent synaptic plasticity.

To understand why L-DOPA cannot fully recapitulate natural dopamine dynamics in people with PD, we must consider its mechanism of action in combination with known patterns of dopamine neuron degeneration in PD. L-DOPA is converted to dopamine by aromatic L-amino acid decarboxylase (AADC) and can be released from intact dopamine terminals, but also from serotonergic terminals (Maeda et al., 2005; Tanaka et al., 1999). Multiple cell types express AADC (Arai et al., 1996; Kitahama et al., 1990), however, and it is unknown what circuit elements take up L-DOPA and release dopamine in advanced PD. Indeed, as degeneration proceeds, fewer and fewer intact dopamine terminals are available to release dopamine in natural patterns, which may drive worsening outcomes. It has long been known that dopamine neurons from different midbrain regions are selectively vulnerable to PD, and that axons terminating in striatal subregions degenerate at different rates (Surmeier et al., 2017). This observation suggests the pattern of dopamine release from converted L-DOPA may change over time, increasingly diverging from naturalistic patterns. As serotonergic neurons or other cell types start to take up and release dopamine, aberrant patterns of release may develop and drive abnormal dopamine-dependent synaptic plasticity. Although the role of serotonergic neurons still remains somewhat unclear, release of dopamine from serotonergic terminals has been linked to the development of L-DOPA-induced dyskinesias in parkinsonian animals treated with levodopa (Carta et al., 2007; Rylander et al., 2010). Additionally, as dopamine neurons slowly degenerate, other adaptations in dopamine signaling machinery and downstream basal ganglia circuit function are likely to influence L-DOPA's efficacy.

The better we understand dopamine system function, both in the intact brain and in degenerative disease models, the better we will be able to develop therapeutics that restore healthy dopamine dynamics at different stages of disease progression. New approaches are required to address disease symptoms that are currently untouched, or even worsened, by dopamine replacement therapy, such as cognitive functions. Early intervention to restore natural dopamine dynamics might also slow disease progression or prolong the efficacy of dopaminergic treatment by preventing cellular, synaptic or circuit adaptations, including aberrant synaptic plasticity, which have been linked to complications such as levodopa-induced dyskinesia (Borgkvist et al., 2018; Picconi et al., 2003).

Here, we review recent conceptual advances in our basic understanding of the dopamine system with special attention to their import for PD. We summarize our understanding of dopamine circuitry across three axes: (1) molecular identity, (2) network identity, and (3) computational identity. We address how each of these identities may relate to PD and L-DOPA efficacy and provide thoughts on paths forward in PD-related dopamine research. Continued integration of dopamine-related basic and preclinical research will be important to stimulate breakthroughs in PD treatment.