14 September 2012. A nagging problem with many Parkinson’s mouse models is their failure to capture neuron loss, a critical feature of this disease. Case in point: People with loss-of-function mutations in the DJ-1 gene (aka PARK7) develop a rare, recessive form of parkinsonism with profound dopaminergic neurodegeneration (Bonifati et al., 2003; Hague et al., 2003), yet DJ-1 knockout mice on various backgrounds have no appreciable cell loss. Curiously, a subset of DJ-1-null mice has now emerged with the desired phenotype—age-dependent, unilateral loss of dopaminergic neurons in PD-affected brain areas, followed by mild motor symptoms. The serendipitous discovery, reported online September 10 in the Proceedings of the National Academy of Sciences USA, came as David Park, University of Ottawa, Canada, and colleagues were backcrossing DJ-1 knockouts onto a C57 wild-type background.
While these mice do model key events in the human disease, they lack α-synuclein pathology—a hallmark of the sporadic form of Parkinson’s that accounts for more than 90 percent of cases. The genetic basis for the new model remains unclear, but further studies with these animals could identify potential modifier genes for the DJ-1-related degenerative phenotype.
The DJ-1 knockout mouse in the present study was developed in the lab of coauthor Tak Mak at the Campbell Family Institute for Breast Cancer Research in Toronto, Canada. “The phenotype was rather depressing,” Park told Alzforum. “The only thing we could really see was the animals were a bit more hypersensitive to exogenous stress.” Dopamine neuron numbers in the striatum were normal, only dipping slightly when they were exposed to the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyrindine (MPTP) toxin (Kim et al., 2005). MPTP induces dopaminergic loss in normal mice and provides the basis for other partial PD animal models.
Wanting to explore the function of DJ-1 in brain neurons, first author Maxime Rousseaux and colleagues did a series of backcrosses to get the DJ-1 knockouts onto a pure C57Bl/6J background. Surprisingly, some of the DJ-1-C57 mice recapitulated the asymmetry seen in people with PD who only lose dopaminergic neurons in one hemisphere of the brain. The cell death “was spectacular,” Park said. “We barely even had to count cells.” Affected mice had 40 percent fewer dopaminergic cells on one side of the substantia nigra than the other.
The model has incomplete penetrance. At two months of age, about 15 percent of DJ-1-C57-null mice show the phenotype. Penetrance peaks around 12 months, when some 35 to 40 percent of DJ-1-C57 knockouts are affected. Furthermore, the degeneration worsened as the mice aged. By 15 months of age, DJ-1-C57 mice were losing neurons in a bilateral fashion—not only in the substantia nigra, but also in another brain area, the locus ceruleus. They also developed mild motor defects.
This is not the first model to show considerable loss of dopaminergic neurons. Exposing mice to chemical toxins can induce their degeneration. However, treatments based on preclinical studies in these models have fared poorly in clinical trials (see Parkinson Study Group PRECEPT Investigators, 2007; Waldmeier et al., 2006; Snow et al., 2010), leading some to question the relevance of acute toxin models to human PD.
The present report describes “a natural model,” Park said. “We don’t artificially induce [the phenotype]. It’s what happens in the mice as they live their life in the cage.” At least one other model seems to recapitulate the characteristic nigral cell death through genetic manipulation alone (see ARF related news story on Kittappa et al., 2007).
Besides the incomplete penetrance of the neurodegeneration phenotype, some say the biggest concern with the current DJ-1-C57 mice is their lack of α-synuclein pathology. Because this is a defining feature of sporadic PD, but not of the rare inherited, recessive forms, “some people question whether the [recessive] cases should even be called PD. They might be a form of parkinsonism that has dopaminergic neuron loss but lacks pathological features,” said Marie-Francoise Chesselet of the University of California, Los Angeles. Hence, the DJ-1-C57 model may not be useful for preclinical testing of PD drugs, but rather “could lead to advances by identifying modifiers of pathology,” Chesselet said. As a first step toward finding such modifiers, the authors did whole-exome sequencing to identify genetic polymorphisms that segregate with the neurodegenerative phenotype. “How these factors regulate dopaminergic loss in DJ-1-deficient mice will require further analysis,” the authors wrote.—Esther Landhuis.
Rousseaux MW, Marcogliese PC, Qu D, Hewitt SJ, Seang S, Kim RH, Slack RS, Schlossmacher MG, Lagace DC, Mak TW, Park DS. Progressive dopaminergic cell loss with unilateral-to-bilateral progression in a genetic model of Parkinson disease. PNAS Early Edition. 10 Sep 2012. Abstract