When it comes to propagating misfolded tau in cells, a small fragment of the protein may be sufficient, at least according to an April 3 report in Nature Chemistry. A 31-residue stretch of the third microtubule binding domain (R3) instigates tau aggregation in cells, report Jan Stöhr, William DeGrado, and colleagues at the University of California, San Francisco. The researchers described how subtle changes in the domain’s secondary structure modulate the process. A hexapeptide at the beginning of the R3 domain rapidly forms amyloid fibers in vitro, but only the larger 31-amino-acid peptide can pull this off in human cells. The findings highlight the importance of the R3 domain in tau misfolding and propagation, and caution against drawing conclusions based solely on fibrils formed in vitro, Stöhr said.

If replicated, the finding also raises the possibility that this small fragment could propagate tau misfolding in the brain. Scientists still don’t know how that occurs. Tsuneya Ikezu of Boston University School of Medicine noted that truncated forms of tau containing the R3 domain have been spotted in cerebrospinal fluid. Ikezu was not involved in the curret work.

Bare Minimum.

A hexapeptide (bold) at the beginning of the R3 domain rapidly forms fibrils in vitro, but the full 31-residue R3 domain is required to coax tau into aggregates in cells. [Image courtesy of Stöhr et al., Nature Chemistry, 2017.]

Misfolding and aggregation of tau is a hallmark of Alzheimer’s disease and other neurodegenerative tauopathies. Variable splicing produces six tau isoforms, each containing a microtubule-binding domain that is repeated either three or four times. The third repeat, shared by all isoforms, is necessary to form fibrils. Specifically, the hexapeptide VQIVYK at the R3 N-terminus coaxes longer forms of tau into paired helical filaments in vitro (see Research TimelineSawaya et al., 2007; Jun 2011 news). How the parameters for fibril formation in a dish match up with those for propagation in cells was unclear, however. 

Stöhr and colleagues recently found that while short regions of the Prion protein, PrP, readily formed fibrils in a cell-free system, longer segments were necessary to form infectious prions (see Wan et al., 2015). They wondered if longer fragments were necessary for tau propagation, as well. “We wanted to figure out at what point aggregation and propagation in cells converge,” Stöhr told Alzforum.

The researchers generated a series of tau peptides, starting with the 31-residue R3 domain, which spans amino acids 306-336, and truncating it five residues at a time from the C-terminus. Outside of cells, the shortest peptides aggregated most readily. The hexapeptide R3311 (amino acids 306-311) rapidly transformed into fibrils, reaching peak aggregation within 40 minutes. The full R3 domain took around 25 days to do the same. Under the electron microscope, fibrils derived from the shortest peptides appeared uniform, while those formed from the entire R3 domain were a mishmash of fibril structures.

In cells, however, the kinetics were quite different. The researchers added the peptides to HEK-293 cells expressing tau-RD-YFP, a fluorescently tagged version of tau containing all four repeat domains, and harboring both the P301L and V337M mutations. Fluorescence of this reporter serves as a proxy for aggregation. While the 31-residue R3 domain pulled tau-RD-YFP into fluorescent clusters, the aggregation power of the peptide dropped about tenfold for each five- or 10-residue truncation. The hexapeptide itself was a complete dud, whereas R3’s seeding capabilities rivaled that of pre-formed tau-RD fibrils. This surprised Stöhr. “These findings suggest that fast aggregation in vitro does not equate to propagation in cells,” he said.

R3 as Seed. The full 31-residue R3 domain (top left), but not shorter peptides, pulled tau-RD into aggregates (yellow) in kidney cells. [Image courtesy of Stöhr et al., Nature Chemistry, 2017.]

Using a variety of structural techniques—including X-ray fiber diffraction, solid state nuclear magnetic resonance, and hydrogen-deuterium exchange—the researchers determined that the 31-residue R3 domain formed a parallel, three-stranded cross-β structure. Fiddling with the sequence or oxidation state of R3 produced subtle changes to the peptide’s secondary structure, which in turn affected its ability to seed tau-RD aggregation in cells.

The findings highlight the ability of the R3 domain alone to seed misfolded tau in cells. They also emphasize that fibril formation and seeding in vivo are not one and the same, Stöhr said. As to why the hexapeptide failed to seed, he speculated that it might require greater structural complexity than six amino acids have to offer. Ikezu agreed, adding that the heterogeneity of fibrils produced by the 31-residue R3 domain may present more opportunities for interactions with tau protein in cells.

Ikezu pointed out that disease-causing mutations such as P301L and V337M lie outside of the R3 domain. So do phosphorylation sites associated with tau misfolding. He speculated that these mutations or modifications may increase propagation of full-length tau by somehow making the R3 domains more exposed to one another.—Jessica Shugart


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Structure-Based Approach Yields Tau Inhibitors

Mutations Citations

  1. MAPT P301L
  2. MAPT V337M (Seattle)

Paper Citations

  1. . Atomic structures of amyloid cross-beta spines reveal varied steric zippers. Nature. 2007 May 24;447(7143):453-7. PubMed.
  2. . Truncated forms of the prion protein PrP demonstrate the need for complexity in prion structure. Prion. 2015;9(5):333-8. Epub 2015 Sep 1 PubMed.

Other Citations

  1. Research Timeline

Further Reading


  1. . β-Sheet core of tau paired helical filaments revealed by solid-state NMR. J Am Chem Soc. 2012 Aug 29;134(34):13982-9. PubMed.

Primary Papers

  1. . A 31-residue peptide induces aggregation of tau's microtubule-binding region in cells. Nat Chem. 2017 Sep;9(9):874-881. Epub 2017 Apr 3 PubMed.