The researchers were not planning to study Parkinson’s, but were overexpressing IFN-γ in transgenic amyloid precursor protein (APP) mice that model Alzheimer’s disease. Chakrabarty injected the cerebral ventricles of two-day-old pups with an adeno-associated virus vector carrying mouse IFN-γ. But she noticed that both the APP mice, as well as the wild-type controls, developed calcifications in the basal ganglia and thalamus upon overexpression of IFN-γ. “That looks like Fahr’s syndrome,” commented coauthor Dennis Dickson of the Mayo Clinic in Jacksonville, Florida, where Chakrabarty and Golde began the work. People with idiopathic basal ganglia calcification commonly present with symptoms of parkinsonism, such as impaired movement and speech. With the disease so rare, Golde said, it is unclear what the range of pathology could be, and thus how closely the mice parallel the human disease.

Chakrabarty used a viral vector carrying GFP as a control in further studies of the IFN-γ in wild-type mice. The animals injected with the IFN-γ vector showed microgliosis and half died within five months. In addition to the calcifications, evident at five and eight months of age in the surviving animals, the IFN-γ led to neurodegeneration in the nigrostriatum. Starting at five months, the animals had 50-84 percent fewer of the nigrostriatal marker proteins phosphatase 1, regulatory (inhibitor) subunit 1B (DARPP32/PPP1R1B), and tyrosine hydroxylase than did normal mice. The nigrostriatal tract was “pretty much obliterated,” Golde said.

“They are not actually talking about Parkinson’s per se,” noted Shawn Hayley of Carleton University in Ottawa, Ontario, Canada. “It is like they are seeing elements of two different pathologies.”

Behavioral symptoms of parkinsonism were also present in the IFN-γ-overexpressing mice. They spent more time resting and took longer to cross a beam than mice treated with a virus carrying only GFP. In a fear test, they also tended to freeze in place more often than controls, mimicking a parkinsonian gait. Some other parkinsonian models are usually based on chemical toxins such as paraquat and MPTP, which destroy the substantia nigra. The new model, Golde said, is the first to cause neurodegeneration by means of an endogenous factor, and it does so with more specificity for the nigrostriatum than other models.

It is that nigrostriatal decay, more than the similarities to any particular disease, that most interests Golde. Why Parkinson’s pathology hits this part of the brain so hard is a longstanding question. “It is a potentially useful model for studying progressive nigrostriatal degeneration,” he said. The effect is also at least somewhat specific to IFN-γ, as overexpressing IL-6, another inflammatory cytokine, did not lead to neurodegeneration.

Another way to test the role of IFN-γ in neurodegeneration would be to knock out the cytokine. Hayley has done just that (Mangano et al., 2011). He found that IFN-γ knockout mice were protected from the toxicity of paraquat. Similarly, IL-6 knockout animals did not get this protective effect. “My hypothesis is that the microglial inflammatory response is critical for how the brain deals with a whole bunch of environmental toxins and infections,” he said. The two studies complement each other nicely, Hayley told ARF, and suggest that inflammation and IFN-γ are worthy of further study in the Parkinson’s field.

Indeed, both studies point to IFN-γ as a killer of dopaminergic neurons, but the “how” remains fuzzy. One possibility, Golde said, is that IFN-γ causes gliosis that somehow damages the nigrostriatal tract. Microglia produce free radicals to kill infections, Hayley noted, and if that went into overdrive, the free radicals might damage neurons, too. However, Golde favors the idea that IFN-γ itself is directly toxic to the nigrostriatal neurons, perhaps due to those cells differentially expressing the IGF-γ receptor or responding differently to the cytokine. It is also unclear how IFN-γ overexpression causes the basal ganglia calcification. The calcium inclusions could be the direct result of IFN-γ, or simply a nonspecific marker for damaged tissue, Golde said.

“This raises the intriguing specter that altered innate immune responses driving high levels of IFN-γ could be a risk factor for Parkinson’s disease later in life,” Golde said. Some scientists have theorized that encephalitis caused by the 1918 influenza pandemic caused later neurological symptoms (Maurizi, 2010), and basal ganglia calcification is linked to juvenile human immunodeficiency virus (HIV) infection (Mahadevan et al., 2008). Transient upregulation of IFN-γ during a brain infection might knock out a fraction of nigrostriatal neurons, predisposing a person to later Parkinson’s disease, Golde suggested.

“Targeted anti-IFN-γ therapy may be a way to delay or attenuate the progressive loss of nigral dopaminergic neurons,” suggested Malú Tansey of the Emory University School of Medicine in an e-mail to ARF. However, Golde and Hayley noted, it is difficult to imagine implementing such a therapy even if the theory about infection, inflammation, and parkinsonism proves correct. The inflammation could precede the appearance of Parkinson’s symptoms by decades, and it is difficult to predict which individual infections involve nigrostriatal damage. However, Hayley said, it would be a good idea to treat a person with anti-inflammatory medication, once the infection subsides, to dampen any raging IFN-γ and potentially protect neurons.—Amber Dance.

Reference:
Chakrabarty P, Ceballos-Diaz C, Lin WL, Beccard A, Jansen-West K, McFarland NR, Janus C, Dickson D, Das P, Golde TE. Interferon-gamma induces progressive nigrostriatal degeneration and basal ganglia calcification. Nat Neurosci. 2011 May 15. Abstract