18 October 2011. In Alzheimer’s and other brain diseases, normal proteins adopt strange conformations that make them clump together and cause trouble for neurons, perhaps even killing them. Two new studies provide insight into the origins of this process. In an October 4 Molecular Psychiatry paper, researchers led by Claudio Soto at the University of Texas Medical School, Houston, report that injecting human AD brain extract brings on brain amyloid deposition in transgenic mice that do not normally develop plaques. The results add to prior data raising the specter of prion-like transmission for some sporadic AD cases. And in last week’s Journal of Neuroscience, Mathias Jucker of the University of Tübingen, Germany, and colleagues propose that the most robust “seeds” of cerebral amyloidosis may be soluble. If these perilous Aβ forms could be identified in bodily fluids, they could potentially serve as diagnostic AD biomarkers or targets for early intervention.
Unlike in classic infectious diseases, the transmissible agents in prion disorders such as bovine spongiform encephalopathy (BSE, aka “mad cow disease”) and Creutzfeldt-Jakob disease are simply normal proteins that change shape to become destructive. AD’s amyloid-β peptides also exhibit this uncanny property, prompting Soto and others to explore whether AD pathology can be triggered in prion-like fashion—that is, by inoculation with the misshapen protein. Prior work from Jucker’s and Lary Walker’s group at Emory University, Atlanta, Georgia, showed that postmortem AD brain extract can hasten amyloid deposition not only when injected into the brains of AD transgenic mice (see ARF related news story on Meyer-Luehmann et al., 2006; Kane et al., 2000), but also when it gets in via the periphery (see ARF related news story on Eisele et al., 2010). However, the host animals in those studies were already predisposed to develop AD pathology because they overexpress mutant amyloid precursor protein (APP); the inoculate simply quickened the process.
In the Molecular Psychiatry study, first author Rodrigo Morales and colleagues wanted to see if they could stimulate Aβ deposition from scratch, that is, in transgenic mice (HuAPPwt) expressing wild-type human APP and showing no evidence of plaques at the ripe age of over 24 months. The researchers gave five-month-old HuAPPwt mice a shot of AD brain extract, or control extract from a young person, into the hippocampus, and checked for Aβ aggregates 285, 450, or 585 days later. As judged by thioflavin S and anti-Aβ antibody (4G8) staining, none of the control mice showed detectable Aβ aggregates, whereas animals receiving AD extract had diffuse Aβ deposits at the first two timepoints and full-blown, ThioS-positive plaques by day 585. The extent of Aβ aggregation and astrogliosis in these mice intensified with age, and showed up in brain areas far from the injection site, suggesting the seeding activity can spread.
While popular press articles tout the findings as evidence that AD could be “contagious” (see CBS News and Fox News coverage), scientists point out that the new and prior studies were done in AD mouse models that do not develop the full spectra of AD symptoms—only Aβ protein aggregates in the brain—even when injected intracerebrally with AD brain extract. “No experiment has yet shown that AD per se can be transmitted in a prion-like fashion,” wrote Walker in an e-mail to ARF. Furthermore, transmission of prion disease to humans is extremely rare, with less than 700 known cases to date, most under extraordinary circumstances such as treatment with contaminated growth hormone from human cadaveric pituitary glands or in the wake of the BSE outbreak in the 1980s and 1990s (Belay and Schonberger, 2005). No case of AD having been transmitted from person to person, or from animal tissue to a person, has ever been described.
Rather, the current paper lends further support to a concept bandied about for some time—the possibility that protein aggregates in prion diseases, AD, and other brain disorders may form and spread by a common molecular mechanism. Walker and Jucker call it “corruptive protein templating” (aka “seeding”), and lay out evidence for this as a “prime mover of the neurodegenerative process” in a forthcoming Annals of Neurology review (currently online as an “accepted article”). Walker also led an ARF Webinar on this topic.
Earlier primate studies suggested that exogenous Aβ protein could trigger cerebral amyloidosis (Baker et al., 1993; Ridley et al., 2006), and Walker has data, recently submitted for publication, showing prion-like induction of Aβ pathology in transgenic rats that express mutant human APP but do not develop plaques or cerebral amyloid angiopathy through 30 months of age (Agca et al., 2008). Evidence for prion-like propagation of protein aggregates is also accumulating for Parkinson’s (see ARF related news story on Desplats et al., 2009; Hansen et al., 2011; Kordower et al., 2011), Huntington’s (see Ren et al., 2009), and amyotrophic lateral sclerosis (Chia et al., 2010; Munch et al., 2011; ARF related news story on Furukawa et al., 2011).
While these studies point to the ability of amyloidogenic proteins to act as seeds, the nature of the seed remains a mystery. In the Journal of Neuroscience paper, Jucker’s team further characterized the Aβ seeding factor in APP transgenic mouse brain. Using four-month-old pre-plaque APP23 transgenic mice as hosts, first author Franziska Langer and colleagues injected their hippocampi with Aβ aggregate-containing extract from aged APP23 brains, and analyzed brain amyloid load in the treated animals four to five months later. They treated the injection material in various ways to correlate its seeding ability with its biochemical properties.
In one set of experiments, the researchers found that proteinase K (PK)-treated extracts could still induce amyloidosis in the host mice, albeit only 55 percent as robustly as untreated inoculate. While this jibes with other work suggesting that increased protease resistance makes misfolded proteins better at seeding, the data also argue that PK-sensitive seeds exist, since PK-treated extract did not trigger amyloid deposition as robustly as non-treated extract.
To examine that possibility, the researchers used ultracentrifugation to spin out the larger aggregates from the aged APP23 extract. When they injected the leftover soluble material into younger APP23 host mice, it was “surprisingly effective at inducing formation of new plaques,” said Walker, a coauthor on the J. Neuroscience paper. Though less than 0.05 percent of the Aβ remained in the supernatant after ultracentrifugation, this soluble fraction accounted for up to 30 percent of the seeding activity and was PK sensitive. Aβ aggregates arising from soluble seeds tended to be smaller and more uniformly distributed, relative to Aβ assemblies induced by the insoluble fraction.
Scientists called this a nice paper with interesting results and carefully executed experiments. However, Sylvain Lesne of the University of Minnesota, Minneapolis, would like to have seen a more rigorous biochemical characterization of the soluble and insoluble fractions. “Only monomeric Aβ levels are reported,” Lesne wrote in an e-mail to ARF. “It would have been more relevant to document the relative content of oligomeric Aβ molecules (e.g., dimers, trimers, Aβ*56, and annular protofibrils) present in all source material used.” (See comment below.) Lesne and others discussed challenges of studying physiologically relevant Aβ oligomers in a Webinar last week (see ARF Webinar).
Marc Diamond of Washington University School of Medicine in St. Louis, Missouri, noted that it may be difficult to extrapolate the findings to AD given the “rather peculiar system—microinjection of aggregates into mice that are ‘primed’ to form aggregates by virtue of overexpressing Aβ.”
For their part, Jucker said his team is forging ahead with studies in AD, pre-AD, and young patients to look for soluble Aβ seeds in body fluids, primarily cerebrospinal fluid.—Esther Landhuis.
Morales R, Duran-Aniotz C, Castilla J, Estrada LD, Soto C. De novo induction of amyloid-β deposition in vivo. Mol Psychiatry. 2011 Oct 4. Abstract
Langer F, Eisele YS, Fritschi SK, Staufenbiel M, Walker LC, Mathias J. Soluble Abeta; Seeds Are Potent Inducers of Cerebral Beta-Amyloid Deposition. J Neurosci. 2011 Oct 12. Abstract
Jucker M and Walker LC. Pathogenic Protein Seeding in Alzheimer's Disease and Other Neurodegenerative Disorders. Ann Neurol. Oct 2011. Abstract