Hanging out with the wrong crowd can make a person go bad, the adage says. This may be true for β amyloid as well. In a Nature paper published online May 2, researchers led by George Bloom at the University of Virginia, Charlottesville, and Hans-Ulrich Demuth at the German biotech company Probiodrug AG, based in Halle/Saale, report that a highly neurotoxic, pyroglutamylated form of Aβ acts as a “bad seed,” corrupting normal peptide. After regular Aβ42 has been exposed to pyroGluAβ, the peptide adopts the toxic properties of the pyroglutamylated form, and can then transmit this neurotoxicity to other Aβ42, similar to the way misfolded prion proteins propagate. The authors demonstrate that pyroGluAβ’s toxicity requires the presence of tau protein, strengthening the idea that tau is a crucial downstream mediator of the negative effects of β amyloid. They also find that pyroGluAβ is present in Aβ oligomers in AD brains but not in oligomers from normal brains, implying this prion-like transmission could be occurring in AD. “I think this means we need to be paying a lot more attention to pyroglutamylated β amyloid as a potential diagnostic and therapeutic target,” Bloom told Alzforum. Some of this research was previously presented at the 2010 Society for Neuroscience conference in San Diego, California (see ARF related news story).

“The biochemical behavior described here seems striking and is scientifically well grounded,” said Dennis Selkoe at Brigham and Women’s Hospital, Boston, Massachusetts, who was not involved in the work. Selkoe noted that the central question now is how important this form of Aβ is in the pathobiology of AD. For example, does it arise early in the disease or only as a late modification?

PyroGluAβ forms when an Aβ peptide loses its first two N-terminal amino acids, exposing a glutamate residue. The enzyme glutaminyl cyclase (QC) then links the free N-terminal to the glutamate’s side chain, forming a ring-shaped pyroglutamate residue. The resulting pyroGluAβ clumps together more readily than do other forms of the peptide, and it poisons neurons even at very low concentrations. Some researchers suggest that this sticky peptide, which by some estimates makes up as much as half of the Aβ in plaques in AD brains, kicks off β amyloid deposition (see ARF related news: AD/PD 2007 Salzburg story; Keystone 2008 story; AD/PD 2009 Prague story; SfN 2009 Chicago story; and SfN 2010 San Diego story). How and why these pyroglutamylated forms are so toxic has been a mystery. In the June Journal of Neurochemistry, researchers led by Demuth report that pyroglutamylated forms of Aβ are more hydrophobic than their unmodified kin, explaining why they glom together more easily. These modified forms also disrupt long-term potentiation (LTP) in hippocampal slices at lower concentrations than the parent peptides do, the authors report.

Now, in the Nature paper, co-first authors Justin Nussbaum at the University of Virginia and Stephan Schilling at Probiodrug also addressed how pyroGluAβ exerts toxicity. Nussbaum and colleagues allowed solutions of synthetic monomeric Aβ42, synthetic pyroglutamylated Aβ3-42, or a 19:1 mixture of the two to oligomerize in vitro for 24 hours, then tested each solution on cultured mouse neurons. At a concentration of 0.5 μM, Aβ42 killed very few neurons, while both the pyroglutamylated peptide and the 5 percent pyroGluAβ mixture eliminated about half of the cells within 24 hours. One central problem in the field is that researchers use different forms and concentrations of Aβ to test for toxicity (see ARF Webinar). Bloom pointed out that this study aimed to find the most sensitive conditions under which β amyloid harms neurons, as this might help pinpoint how the earliest effects start in the human brain. He told Alzforum that Aβ42 did not become cytotoxic until it reached concentrations of 10 or 20 μM.

A related problem concerns how synthetic Aβ peptides made by one protocol compare to those made by another, or to native peptides from AD brain. All of the pyroglutamylated peptides used in these studies were made at Probiodrug using the company’s own protocols, Bloom said. For Aβ42, his lab switched between using Probiodrug peptides and those made with a different protocol by coauthor Charlie Glabe at the University of California, Irvine. They saw no difference in the experimental effects of the two varieties, Bloom reported. Demuth told Alzforum that his company has compared its synthetic peptides to those produced in cell culture, and found that both kinds similarly affect LTP, as described in the Journal of Neurochemistry paper.

What, then, explains the relative toxicities of the different preparations of Aβ? Under the oligomerization conditions used, pure Aβ42 started to aggregate into pre-fibrils after three hours, pure pyroGluAβ stayed dimeric or trimeric for 24 hours, and dimers and trimers in the mixture lasted for up to three days. This is significant, because small oligomeric forms of Aβ are now widely believed to be the most toxic variety. By fractionating the solutions with gel filtration, the authors confirmed that most of the cytotoxicity in their pyroGluAβ and mixed solutions came from the dimeric or trimeric forms. Using co-immunoprecipitation, Nussbaum and colleagues found that pyroGluAβ and normal Aβ42 formed hybrid oligomers in the mixture.

The varying toxicities and kinetics of oligomerization of the different Aβ species imply that pyroGluAβ oligomers assume a different shape than oligomers of Aβ42, and can pass that template on to normal Aβ42, the authors note. What this toxic structure might be is a mystery. Bloom told Alzforum he plans to pursue this question through collaboration with biophysicists and structural chemists. To see how persistent the toxic conformation is, the authors serially diluted the 5 percent pyroGluAβ/95 percent Aβ42 mixture to the point where pyroGluAβ represented one in 160,000 of the total Aβ. The resulting solution was nearly as cytotoxic as the original mixture, suggesting a prion-like transmission mechanism. In future work, Bloom said, he plans to add other forms of Aβ, such as Aβ40, to the mixture to see how that affects toxicity. He noted that such complex mixtures would more closely mimic the brain environment.

The brain environment contains plenty of tau, and recent work by several groups has shown that tau acts downstream of β amyloid (see ARF related news story; ARF news story; and ARF related news story). To investigate the connection between pyroGluAβ and tau, Schilling and colleagues at Probiodrug used the company’s TBA2.1 transgenic mice, which express truncated Aβ in the brain, accumulate small amounts of pyroGluAβ (less than 100 ng per gram of brain weight), and develop massive neurodegeneration and gliosis by three months of age (see Alexandru et al., 2011). When Schilling and colleagues crossed these mice with tau knockout animals, the double transgenics were almost completely protected from degeneration and gliosis. As Bloom put it: “I think of amyloid and tau as being the trigger and bullet. Amyloid somehow initiates a pathogenic cascade, but most outputs of that cascade require tau.”

The authors suggest that these findings may relate directly to human AD. Using antibodies specific for Aβ42 oligomers and pyroGluAβ, and brain extracts supplied by Glabe, Nussbaum, and colleagues found that the pyroglutamylated form lurked in Aβ oligomers in three out of three AD brains, but was absent or barely detectable in three healthy brains. Probiodrug is developing a therapy based on inhibiting the QC enzyme that makes pyroGluAβ. Their lead compound, PQ912, has completed Phase 1 trials and proven safe over a concentration range of three orders of magnitude, Demuth told Alzforum (see company press release). These trials included a single ascending dose and multiple ascending dose study in healthy young volunteers, as well as a bridging trial in elderly volunteers. The drug pharmacokinetics were more favorable in the elderly participants, Demuth said, with better brain delivery and properties. He will present the data at the 12th International Stockholm/Springfield Symposium on Advances in Alzheimer’s Therapy, held 9-12 May 2012 in Stockholm, Sweden.

Demuth said the company is currently preparing patient studies. One of the questions is whom to recruit. To help answer this, Probiodrug is gathering blood and cerebrospinal fluid samples from patients and normal aged controls, and comparing their Aβ oligomer profile and behavioral symptoms. Based on animal studies, Demuth said, “data suggest an early application of this approach would be the best.” This mirrors a trend in the field to start amyloid-targeting therapies at very early disease stages (see ARF Webinar).—Madolyn Bowman Rogers


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. Nussbaum et al. show that when as little as 5 percent of Aβ is in the pyroglutamylated N-truncated form (Aβ 3(pE)-42), it confers toxicity to a solution of Aβ1-42 monomers. The authors suggest that Aβ42 undergoes prion-like templated conversion induced by the 3(pE)-42 seed. Given that Aβ aggregation is a self-propagating phenomenon, even a minute amount of seed may be sufficient to drive amplification of toxic Aβ aggregates.

    It is also interesting that the authors carefully control time, buffer, and concentrations, as all these factors influence the final equilibrium between toxic and non-toxic species in the Aβ mixture. The principle that small alterations in Aβ species have strong effects on the final biological and toxic effects of Aβ is, however, not novel (see, e.g., Kuperstein et al., 2010, and Pauwels et al., 2011). As a note of caution, it is not very clear why the authors did not find any pathogenic effects of Aβ42 at low micromolar concentration, and what would happen if they used more abundant Aβ40 instead of Aβ42 as a substrate for seeding. The authors also do not comment on other amyloid fragments that may serve as templates and induce toxic conformations (Aβ43, Saito et al., 2011; phosphorylated Aβ, Kumar et al., 2011; different kind of oligomers, and so on).

    Intriguingly, the authors show that the hybrid pE-Aβ oligomers may exist in vivo (though patient number was quite limited), but do not proceed with further characterization of these species. Nevertheless, it is an interesting work that potentially provides new insights into amyloid toxicity in AD.


    . Neurotoxicity of Alzheimer's disease Aβ peptides is induced by small changes in the Aβ42 to Aβ40 ratio. EMBO J. 2010 Oct 6;29(19):3408-20. PubMed.

    . Structural basis for increased toxicity of pathological aβ42:aβ40 ratios in Alzheimer disease. J Biol Chem. 2012 Feb 17;287(8):5650-60. PubMed.

    . Potent amyloidogenicity and pathogenicity of Aβ43. Nat Neurosci. 2011 Aug;14(8):1023-32. PubMed.

    . Extracellular phosphorylation of the amyloid β-peptide promotes formation of toxic aggregates during the pathogenesis of Alzheimer's disease. EMBO J. 2011 Jun 1;30(11):2255-65. PubMed.


News Citations

  1. San Diego: Pilin’ on the Pyro, Aβ Going Rogue
  2. Salzburg: Aβ’s N-terminal Shenanigans
  3. Keystone Drug News: Pyroglu Aβ—Snowball That Touches Off Avalanche?
  4. Prague: Piecing Together Pathology with PyroGlu
  5. Chicago: Interest in PyroGluAβ Flares Up in Academia
  6. The Plot Thickens: The Complicated Relationship of Tau and Aβ
  7. Honolulu: The Missing Link? Tau Mediates Aβ Toxicity at Synapse
  8. APP Mice: Losing Tau Solves Their Memory Problems

Webinar Citations

  1. Clearing the Fog Around Aβ Oligomers
  2. Treating Before Symptoms—ADCS Invites Ideas for Clinical Trials in Very Early AD

Paper Citations

  1. . Selective hippocampal neurodegeneration in transgenic mice expressing small amounts of truncated Aβ is induced by pyroglutamate-Aβ formation. J Neurosci. 2011 Sep 7;31(36):12790-801. PubMed.

External Citations

  1. company press release

Further Reading

Primary Papers

  1. . Prion-like behaviour and tau-dependent cytotoxicity of pyroglutamylated amyloid-β. Nature. 2012 May 31;485(7400):651-5. PubMed.
  2. . N-Terminal pyroglutamate formation of Aβ38 and Aβ40 enforces oligomer formation and potency to disrupt hippocampal long-term potentiation. J Neurochem. 2012 Jun;121(5):774-84. PubMed.