Protein clumps are anatomical calling cards of many neurodegenerative disorders. Alzheimer's disease has amyloid plaques and tau tangles, and another protein, alpha-synuclein, can clump in several neurodegenerative disorders, including Parkinson's disease and Lewy body dementia.

But in many ways, the relationship of such clumps to clinical symptoms is still mysterious.

There is the well-known finding that most aged individuals have amyloid plaques in their brains, Alzheimer's or no. And cognitive symptoms can vary widely between individuals who turn out to have roughly the same plaque burden.

Additionally, sometimes different proteins can clump together – alpha-synuclein and tau can be found in the same cells.

Or not.

"You have . . . situations where two different proteins are aggregating," Virginia Lee, of the University of Pennsylvania, told BioWorld Today. "Why is that the case?" And why sometimes, and sometimes not?

In studies published in the July 3, 2013, issue of Cell, Lee and her colleagues provided a partial answer to that question. Alpha-synuclein can seed tau aggregates – but only under some circumstances, when it is folded in the right way to do so.

Protein aggregates are seeded by misfolded proteins, which can transmit disease. The most famous example is Creutzfeldt-Jakob disease. But it is increasingly recognized that other proteins besides prions may work in the same way.

In their studies, Lee and her team looked at two different alpha-synuclein strains to see what their effects on tau would be. Unexpectedly, they found that one strain was able to promote tau clumping, while the other was not.

In a further twist – or fold – to the story, alpha-synuclein could, after repeated rounds of seeding, acquire the ability to induce tau clumping.

The team also looked at brain tissue from five patients who had either only Parkinson's disease, or both Parkinson's and Alzheimer's diseases. They found that the patients with mixed symptoms appeared to have two distinct forms of alpha-synuclein, while those with relatively "pure" Parkinson's disease had only one form of the protein. Although the authors wrote in their paper that "the sample size at this stage is too small to correlate strain conformations with pathological manifestation," the data do confirm the existence of different forms of synuclein in patients with neurodegenerative disease.

"What we've tried to do," Lee explained, "is to come up with a molecular mechanism to explain the heterogeneity of Parkinson's disease and Alzheimer's disease," as part of a larger goal of finding out whether transmissibility is a common feature of neurodegenerative disorders.

The findings suggested that one reason for that heterogeneity is that some patients have, or develop, variants of alpha-synuclein that interact with tau, bringing their situation from bad to worse. Because the two alpha-synuclein strains do not differ in their amino acid sequences, and the alpha-synuclein can acquire the ability to see tau over time without any changes to its amino acid structure, one possibility is that the proteins' shapes are altered as they are transmitted from one cell to another.

Lead author Jing Guo told BioWorld Today that from a practical perspective, the work suggested that "immunotherapy may be a good way of treating this disease," because immune cells are more likely to be able to recognize a protein even when its shape has changed.

The findings also serve as a reminder that how a protein folds may be predictable from its amino acid sequence, but that sequence is not the only thing that determines folding.

"A protein can change its conformation depending on the environment," Lee said, and Guo added that "small changes in structure can lead to very large changes in function."

Together, those two facts add up to a simple but important reminder for drug discovery efforts, Guo said. When targeting proteins in neurodegenerative disease – and quite possibly in other situations, too – "make sure you know which shape you want to go after."