Toxicity Of Isoquinolines
Inhibition of the Respiratory Chain
Neurotoxins may disrupt a variety of cellular activities such as DNA repair, protein trafficking, and neurotrans-
mitter release. Cell death precipitated by neurotoxin inhibition of oxidative phosphorylation may, however, be the mechanism most relevant to PD. With this perspective, McKnaught and colleagues6 examined the effects of three classes of isoquinoline derivatives (e.g., isoquinolines, dihydroisoquinolines, TIQs) or MPP+ on respiratory chain complexes I, II-III, and IV from rat brain mitochondria. The isoquinoline derivatives or MPP+ were added to reaction mixes at a concentration of 10 mM and complexes I, II-III, and IV activities were compared to control. Under the experimental conditions of their study, MPP+ produced approximately 70% inhibition of complex I but had no effects on complexes II-III and IV. At 10 mM, isoquinoline and isoquinolinium ion were associated with complete inhibition of complex I, whereas the effects of TIQ (20%), N-methyl-TIQ (65%), (R/S) salsolinol (55%), and N-methyl-(R/S) salsolinol (60%) were less dramatic. The IC50 values for N-methyl-TIQ (4.3 mM) and N-methyl-isoquinolinium ion (1.3 mM) reported by McNaught et al.6 were similar to those provided earlier by Suzuki and co-workers.52 For comparison, the complex I IC50 for MPP+ is about 4.1 mM.6 Only isoquinoline, 6,7-dimethox-yisoquinoline and N-methyl-(R/S)salsolinol produced notable inhibition of complex II-III. None of the isoquin-oline derivatives inhibited complex IV. For the series of compounds examined, there was no significant general relationship between a lipophilicity measure and the extent of complex I inhibition.
In Vitro Toxicity
The in vitro toxicity of TIQ derivatives is summarized in Table 9.3. Like MPTP,53 several TIQs that have been detected in brain inhibit tyrosine hydroxylase (TH).15,54 In a comparison of six TIQs (e.g., N-methyl-TIQ, N-methyl-IQ+, (R/S)salsolinol, N-methyl-(R/S)salsolinol, norsalsoli-nol, and N-methyl-norsalsolinol), N-methyl-norsalsolinol was the most potent in reducing TH activity.15 TH enzyme activity was virtually eliminated by 1 mM concentrations of N-methyl-norsalsolinol, N-methyl-IQ+, and N-methyl-salsolinol. Scholz et al.54 showed that the inhibition of TH by N-methyl-norsalsolinol was noncompetitive with an IC50 of 10 |M. Therefore, investigators of nigral neuro-toxins should bear in mind that reductions in dopamine content may be a manifestation of TH inhibition rather than an indication of dopaminergic cell death.
Early cell culture studies have provided convincing evidence of isoquinoline neurotoxicity.22,23 More recent studies have explored mechanisms of dopaminergic cell death produced by TIQs using both molecular and morphological assays. In general, N-methylated derivatives are more toxic than their nonmethylated parent compounds, and (R)-enantiomers are more toxic than (S)-enantiomers. Among the TIQ derivates, N-methyl-(R)sal-solinol may be the most potent dopaminergic neurotoxin and produces cell death through apoptotic mechanisms. Using an assay of DNA fragmentation, the (R)-enanti-omer of N-methyl-(R/S)salsolinol is at least tenfold more toxic than the (S)-enantiomer.55 In addition to DNA fragmentation, apoptosis has been demonstrated with TUNEL labeling,55 Hoechst 33342 staining,56 and Western blot detection of caspase-3 activation.56 Dopaminergic cells (SH-SY5Y) overexpressing Bcl-2 are resistant to apopto-sis induced by N-methyl(R)salsolinol.57 In addition, the MAO inhibitors deprenyl and rasagiline prevent cell death produced by N-methyl(R)salsolinol.18,57
In Vivo Neurotoxicity
In vivo neurotoxicological studies of TIQ and its derivatives have failed to provide convincing evidence of cell death within the SNpc; this failure may be, in part, due to methodological issues. For example, no publication to date has employed rigorous stereological counting methods and highly sensitive markers of neurodegeneration such as Fluoro-Jade® B and amino cupric silver staining after challenge with TIQ or its derivatives. Furthermore, small sample sizes, variable routes of compound administration, and wide-ranging dosing durations compromise any effort to summarize the literature on TIQ neurotoxic-ity. Moreover, few studies of TIQ in vivo neurotoxicity have combined morphological, behavioral, and biochemical endpoints into a cohesive analysis.
Two early studies in marmosets suggest that the both the behavioral and biochemical effects of TIQs could be due to TH inhibition rather than cell death within the SNpc. Nagatsu and Yoshida12 injected marmosets with TIQ subcutaneously for 11 days. This dosing regimen was associated with parkinsonism, reduced striatal dopamine, and reduced TH activity. However, they did not describe cell death within the SNpc. In squirrel monkeys, TIQ was injected subcutaneously for up to 104 days at a dosage of 20 mg/kg-day.3 These monkeys exhibited rigidity and bradykinesia, reduced nigral dopamine, and decreased TH activity. Again, there was no indication of anatomical changes within the substantia nigra.
Nigral cell death was also not reported in two investigations of TIQ neurotoxicity in mice;13,58 TIQ was injected subcutaneously on a daily basis for at least four weeks in both studies. In C57BL/6J mice, reduced TH immunoreactivity in the absence of overt cell death was detected.13 Surprisingly, there was no reduction in striatal dopamine, DOPAC or HVA in C57BL mice.58
The N-methylated derivative of TIQ, N-methyl-TIQ, and its oxidized product, N-methyl-IQ+, have as well been examined for in vivo neurotoxicity. In a study by Perry and colleagues,59 after N-methyl-TIQ was injected into marmosets subcutaneously, three out of four animals died; disappointingly, postmortem pathological findings were not reported. The one surviving marmoset showed