The saga of harmful administration of levodopa (LD) in the treatment of Parkinson's disease (PD) resulted from outcomes of animal trials and cell culture studies. They were initiated after the clinical observation of onset of motor complications related to the short plasma half-life of the drug in PD patients. This discussion only partially considered a further aspect, which is associated with the long-term administration of LD. Chronic LD intake increases homocysteine plasma levels. This may support progression of the disease due to concomitant onset of neuropsychiatric symptoms and comorbidities (i.e., vascular disease). In the periphery, therapeutic approaches for this LD-mediated homocysteine increase are vitamin supplementation (i.e., folic acid or application of LD with an inhibitor of catechol-O-methyltransferase [COMT]). In the brain, a blood–brain trespassing precursor of folic acid or a centrally acting COMT inhibitor may represent hypothetical therapeutic approaches. This COMT inhibitor should be applied together with an oxidative stress reducing monoamine oxidase-B inhibitor, in order to force central dopamine metabolism further down via the methylation path. However, this may turn out to be a double-edged sword, since the inhibition of O-methylation with the COMT inhibitor may hypothetically contribute to increased N-methylation. Thus, endogenous tetrahydroisoquinolines may be transformed to neurotoxic N-methylated tetrahydroisoquinolines. These neurotoxic compounds were observed in cerebrospinal fluid and plasma of long-term LD-treated PD patients. They have a structure similar to 1-methyl 4-phenyl 1,2,3,6-tetrahydropyridine or its ion 1-methyl-4-phenylpyridinium, both of which are known to induce PD-like motor symptoms.