AUTHOR=Christensen Christian , Þorsteinsson Haraldur , Maier Valerie Helene , Karlsson Karl Ægir TITLE=Multi-parameter Behavioral Phenotyping of the MPP+ Model of Parkinson’s Disease in Zebrafish JOURNAL=Frontiers in Behavioral Neuroscience VOLUME=14 YEAR=2020 URL=https://www.frontiersin.org/journals/behavioral-neuroscience/articles/10.3389/fnbeh.2020.623924 DOI=10.3389/fnbeh.2020.623924 ISSN=1662-5153 ABSTRACT=

Parkinson’s disease (PD) has been modeled in several animal species using the neurotoxins 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and its oxidized product 1-methyl-4-phenylpyridinium (MPP+). MPP+ selectively kills dopaminergic neurons in pars compacta of the substantia nigra, inducing parkinsonian symptoms in animals. Typically, neurotoxicity models of PD in zebrafish assess acute drug effects on locomotion. In the present study, we examined the lasting effects of MPP+ exposure and drug treatment in zebrafish larvae. Larvae were incubated in 500 μM MPP+, from 1 to 5 days post fertilization (dpf), followed by 24 h drug-free acclimation. At 6 dpf, the behavior was analyzed for locomotion, thigmotaxis, and sleep. Next, in separate assays we assessed the drug effects of brain injected glial cell-derived neurotrophic factor (GDNF) and 4-phenylbutyrate (PBA), co-incubated with MPP+. We show that MPP+ exposure consistently reduces swim distance, movement frequency, and cumulative time of movement; thus mimicking a parkinsonian phenotype of reduced movement. In contrast, MPP+ exposed larvae demonstrate reduced anxiety-like behavior and exhibit a sleep phenotype inconsistent with human PD: the larvae display longer sleep bouts, less sleep fragmentation, and more sleep. Previously reported rescuing effects of PBA were not replicated in this study. Moreover, whereas GDNF attenuated the sleep phenotype induced by MPP+, PBA augmented it. The current data suggest that MPP+ exposure generates a multifaceted phenotype in zebrafish and highlights that analyzing a narrow window of data can reveal effects that may be inconsistent with longer multi-parameter approaches. It further indicates that the model generally captures motor symptoms more faithfully than non-motor symptoms.