12 July 2011. Heiko Braak’s and Kelly Del Tredici’s staging scheme for Parkinson’s disease (PD) raised an intriguing question. Could PD start in the belly and spread from there, along paths laid out by neurons, to successive areas of the brain in a predictable sequence (see Part 1 of this series)? How this might possibly work has become the subject of intense investigation.

The two neuroanatomists at the University of Ulm, Germany, posed the question based on the distribution pattern of Lewy bodies, microscopic protein aggregates containing the protein α-synuclein, in the nervous system of autopsied PD patients. α-synuclein is normally a soluble protein in neurons, but something about the disease makes it fold into the wrong three-dimensional structure, causing it to clump together.

In 2003, Braak and Del Tredici suggested that this disease process is set off “by a yet-unidentified pathogen that is capable of passing the mucosal barrier of the gastrointestinal tract and, via post-ganglionic enteric neurons, entering the central nervous system” (Braak et al., 2003). There is no evidence that PD is an infectious disease. Even so, viruses have been implicated in a number of neurological diseases, and some indirect clues do suggest that a virus might play a role in PD. For example, although controversial, epidemiological data have linked the development of post-encephalitic parkinsonism to the 1918 “Spanish flu” outbreak (Ravenholt and Foege, 1982).

To explore this connection, Richard Smeyne of St. Jude Children’s Hospital in Memphis, Tennessee, and colleagues inoculated mice with a strain of the H5N1 bird flu virus via the nose and showed that the virus can move through the gut and the brain and induce pathological changes that are reminiscent of those seen in PD. These included functional loss of dopamine in neurons in the substantia nigra and an increase in α-synuclein expression and aggregation (see ARF related news story on Jang et al., 2009). “The pattern of infection starting in the enteric nervous system and traveling to the central nervous system via the vagus nerve parallels the progression of PD as described by Braak,” Smeyne said.

One idea that is gaining momentum is that, instead of a virus or some other agent spreading through the nervous system to induce the pathology associated with PD, the pathology itself, in other words, the misfolded form of α-synuclein, is what spreads from one neuron to the next (see image).

The identification of Lewy bodies in nerves leading from the peripheral to the central nervous system has led to the hypothesis that Parkinson’s pathology spreads via transneuronal or trans-synaptic transmission. Image credit: H. Braak and K. Del Tredici

Several studies in vitro and in vivo have indicated that the soluble form of α-synuclein is released at synapses and taken up by other neurons. One of the first of those studies, reported by Seung-Jae Lee, now at Konkuk University in Seoul, South Korea, dates back to around the time Braak first proposed his hypothesis (Lee et al., 2005). “The general reaction of the PD community at the time was understandably not very enthusiastic. Many were probably hesitant to believe our results, and even if they did, they did not know what to make of them,” wrote Lee in an e-mail to ARF. “We learned about Braak’s staging for synucleinopathy around the same time. It was not as well known or well received by the research community as it is today. But when I first read about Braak's hypothesis, I immediately realized the potential implications of α-synuclein release. So I formulated my own hypothesis of α-synuclein aggregates spreading through cell-to-cell transfer.”

Recent research supports Lee’s hypothesis. For example, Pamela McLean’s group at Harvard Medical School in Boston has shown that α-synuclein oligomers are taken up by neurons in culture, transported through the axon to the cell body, and, once there, cause cell death. Moreover, the heat shock protein Hsp70—a chaperone that assists in protein folding and is produced in higher amounts when cells come under toxic or inflammatory stress—reduces the formation of α-synuclein oligomers and related toxicity (Danzer et al., 2011). Together with studies from other groups (see ARF related news story on Desplats et al., 2009), this work raises the possibility that some neurons might transmit a specific conformation of α-synuclein, possibly induced by inflammation or some other mechanism, that then seeds the aggregation of α-synuclein in adjacent neurons. “The latest experimental results are very exciting and tend to point to disease progression via transneuronal or trans-synaptic transmission,” Braak wrote in an e-mail to ARF.

One of the most compelling indications of this proposed spread comes from examinations of PD patients who had received transplants of fetal nigral dopaminergic nerve cells and died several years after the surgery. In 2008, the laboratories of Jeffrey Kordower at Rush University Medical Center in Chicago, Illinois; Patrik Brundin at Lund University in Sweden; and Ole Isacson at Harvard Medical School in Boston independently described results from the autopsies of eight such patients. Two of the groups identified three patients who had Lewy bodies in the transplanted fetal tissues (see ARF related news story on Mendez et al., 2008; Kordower et al., 2008; Li et al., 2008). “The likelihood of dopaminergic nigral neurons developing Lewy pathology at such an age in their native surroundings is exceedingly small. It can be inferred, therefore, that the lesions resulted from more than a decade of interaction between the host tissue and the transplanted nerve cells,” wrote Braak and Del Tredici in an accompanying commentary (Braak and Del Tredici, 2008).

The Lewy bodies showed up in the transplants of patients who had survived 10 years or more after transplantation, but not in people whose transplants were fewer than 10 years old, implying that it takes about this long for α-synuclein pathology to spread. Since those original studies, additional patients with α-synuclein pathology in their transplants have been identified, bringing the total number of such patients to eight, according to Brundin. The most recent case was published on May 18, 2011, in the Journal of Parkinson’s Disease.

Again, how the pathology got to the transplants is subject to debate. “There is no question that they found Lewy bodies in transplants,” said Robert Burke of Columbia University in New York City. “But you cannot draw the conclusion that the α-synucleinopathy was transmitted to these cells. There could be many other explanations. One is that, if you put cells in a toxic environment such as the brain of a PD patient, they can develop aggregates.”

To distinguish among possibilities, researchers have turned to animal models. Brundin’s group grafted wild-type mouse embryonic mesencephalic neurons into the brains of transgenic mice that had been engineered to produce large amounts of human α-synuclein. When they examined brain tissues from these mice, they detected human α-synuclein in the transplanted tissues from the wild-type mice. Thus, the human α-synuclein in the brain cells of the transgenic mice had entered the nerve cells in the transplants (Hansen et al., 2011).

Kordower’s group came to similar conclusions using a different model. They grafted fetal rat brain tissues into the brains of adult rats that had been treated with a toxic chemical known to induce α-synuclein misfolding and aggregation. One month after the transplant, the researchers injected viruses containing the human α-synuclein gene into the rats’ brains at a sufficient distance from the grafts so that none of the graft cells would be injected. When they later examined brain tissues from these rats, they found that a small number of grafted neurons expressed the human α-synuclein, and in some cells the protein was misfolded and aggregated (Kordower et al., 2011). “Clearly, propagation of α-synuclein can occur,” said Kordower. “Whether it does occur in patients remains to be proven.”

Prion diseases, like Creutzfeldt-Jakob disease in people or bovine spongiform encephalopathy and scrapie in animals, are infectious, meaning that they can spread from one person or one animal to another. That has never been shown to be the case for PD. But the studies by Lee, McLean, Kordower, and Brundin suggest that PD might share some characteristics with prion diseases in terms of how α-synuclein spreads from neuron to neuron, and under some circumstances, causes other α-synuclein molecules to misfold and stick together. This line of investigation parallels studies of aggregate migration of the amyloid-β peptide (see ARF related news story on Eisele et al., 2009) and tau (see ARF related news story on Clavaguera et al., 2009; and ARF related news story on Frost et al., 2009).

“For me, this is extremely exciting. Four years ago we would not be talking about this,” said Brundin. “The Braak hypothesis got us talking about a virus. But the actual protein that is transferring is a new concept that we are now investigating.”—Laura Bonetta.

This concludes a three-part series. See also Part 1 and Part 2. Read the entire series.

Laura Bonetta is a freelance writer in Garrett Park, Maryland.