As anyone who has rinsed off after frolicing on a beach knows, tiny particles weasel their way into tight spaces and are far more troublesome than larger ones. The same seems true of soluble aggregates of Aβ, according to a study published January 4 in the Journal of Neuroscience. Researchers led by Dennis Selkoe and Dominic Walsh at Brigham and Women’s Hospital in Boston reported that large soluble oligomers make up the vast majority of soluble Aβ aggregates in the AD brain, yet have minimal cytotoxic power. However, when the researchers let those larger aggregates dissociate, they liberated tiny Aβ clusters that packed an outsized punch—dampening synaptic activity in hippocampal neurons and instigating inflammatory microglial responses in the mouse brain. Conditions that destabilize high molecular weight oligomers or prevent the sequestration of smaller ones could contribute to Aβ toxicity, the researchers concluded.

By virtue of their size, and possibly other biochemical characteristics, these smallest aggregates may cause the most damage by deftly squeezing into cramped spaces like synapses and interacting with cellular membranes, Selkoe told Alzforum. “The smaller an aggregate is, the more likely it is to cause trouble,” he said.

Most researchers agree that soluble Aβ oligomers can exact a toxic toll on neurons, but efforts to examine these pesky Aβ species have been complicated by the very attributes that might make them so toxic: The hydrophobic aggregates readily cling to plastic tubes and pipettes, and interact with each other to shape-shift. Recently, researchers led by David Brody and colleagues at Washington University in St. Louis developed a detailed protocol to isolate soluble aggregates of Aβ from postmortem AD brain extracts, and reported that a majority of aggregates were of high molecular weight (see May 2016 news). However, previous studies led by Selkoe and colleagues reported that while scarce, small oligomers—dimers in particular—caused the most cellular damage (see Jun 2008 newsNov 2007 conference news). 

For the current study, first author Ting Yang and colleagues wanted to directly compare the toxicity of small and large oligomers and determine whether the latter serve as a reservoir for tinier, more noxious species. The researchers prepared postmortem brain extracts from the frontal cortices of six AD patients and five non-AD controls. After a centrifugation to remove insoluble and membrane-associated material, they obtained a mixture containing soluble Aβ species that they then probed with antibodies specific for either monomeric (mAβ 266) or oligomeric (mAβ 1C22) Aβ. They found that in extracts from AD patients, the majority of Aβ existed as oligomeric species. In contrast, monomeric Aβ dominated extracts from healthy controls.

To assess the size of oligomers, the researchers ran the AD brain extracts on size-exclusion chromatography columns followed by western blotting. They found that most soluble Aβ weighed between 150 kDa and 600 kDa. Denaturing these larger conglomerates with detergent yielded both monomers (~4 kDa) and dimers (~7.5 kDa). Yang and colleagues next asked whether dissociating the larger oligomers in non-denaturing conditions might yield smaller oligomers that could contain neurotoxic species. They eluted larger oligomers (greater than 150 kDa) from the size-exclusion column using an alkaline buffer, and left them to marinate in this basic solution at 37° for two days. They found that the alkaline bath caused most oligomers to dissociate, raising levels of monomers and dimers. While a substantial amount of total Aβ was lost between the pre- and post-incubation samples, Selkoe told Alzforum this was at least partly due to oligomers’ infamous stickiness to tubes and pipettes.

Which size oligomer would most aggravate cells in the brain? The researchers first addressed this question by adding the larger oligomers (>150 kDa), either with or without a two-day dissociation in alkaline buffer, to hippocampal slice cultures to measure effects on long-term potentiation (LTP), a cardinal proxy of neural plasticity. The large oligomers had minimal effects on LTP, but the dissociated, smaller oligomers suppressed it. On average, neurons exposed to large oligomers prepared from any of the six AD brain samples had a 48 percent potentiation of excitatory postsynaptic potential in response to electrical stimulation, while neurons exposed to small oligomers mustered only a 29 percent uptick. LTP was unimpeded when the researchers immunodepleted Aβ first. Small, but not large, oligomers also reduced the expression of β2-adrenergic receptor on the neuronal surface by 40 percent, in keeping with previous reports that oligomers trigger the internalization and degradation of these and other receptors (see Wang et al., 2011; Li et al., 2013). 

Selkoe and colleagues had previously reported that Aβ oligomers activated microglia and ramped up their expression of pro-inflammatory cytokines, but what size of oligomers did this was unknown (see Xu et al., 2016). To address this, the researchers injected either large or dissociated/small oligomer mixtures from AD brain extracts directly into the ventricles of healthy, wild-type mice, then examined the activation state of the microglia in the hippocampal and perihippocampal regions after two days. Compared to microglia in mice injected with large oligomers, those in animals treated with small oligomers were rounder and had fewer branches, and expressed higher levels of the activation marker CD68. The researchers concluded that small oligomers fired up microglia, which could trigger damaging neuroinflammatory responses in the brain.

Tiny Terrors. Microglia (expressing P2ry12, red) in the cornu ammonis (top panel) and the dentate gyrus (bottom panel) of the hippocampus round up retract their processes, and express CD68 (white) in response to small, dissociated oligomers (middle). They did not respond to large oligomers (left), or extracts of immonodepleted of Aβ (right). [Courtesy of Yang et al., The Journal of Neuroscience, 2017.]

Overall, the findings paint small oligomers, not their larger forebears, as cerebral agitators, Selkoe told Alzforum. “They also support the conclusion that a huge proportion of soluble Aβ oligomers have no bioactivity,” he added. He thinks a continuous release of small oligomers from larger forms explains why oligomer-specific antibodies detect Aβ around plaques, and why synapses wither in their vicinity (see Feb 2009 news). Selkoe believes that the Aβ dimers released by detergent or alkali represent a particularly toxic, covalently bonded form of oligomer.

Lars Lannfelt of Uppsala University in Sweden questioned how the dissociation of large oligomers in alkaline solution would relate to conditions in the AD brain. “During 48 hours at 37°C, many things can happen, including protease activity changing the nature of the oligomers,” he wrote in a joint comment with Dag Sehlin of Uppsala and Pär Gellerfors, Hanna Laudon, and Linda Söderberg of BioArctic in Stockholm. Selkoe acknowledged that these dissociation conditions are not entirely physiological, however he speculated that similar dissociation likely happens under neutral pH conditions in the brain, albeit over a much longer timescale.

Dietmar Thal of KU Leuven in Belgium said that artificial conditions are a necessary evil in any study making use of postmortem brain extracts or modeling potentially pathological events, but that Yang and colleagues did an excellent job at controlling them. He added that the preponderance of large oligomers in the AD brain agrees with findings from his previous study, which used different methods to come to the same conclusion (see Upadhaya et al., 2011). Lannfelt’s previous work also found a preponderance of large oligomers in the brain. However, using synthetic oligomers, Lannfelt and colleagues came to the conclusion that both large and small species were cytotoxic (see Sehlin et al., 2012). 

Thal added that it will be crucial to address how the dissociation of these large oligomers into smaller, toxic forms is influenced by other proteins in the brain, and by post-translational modifications of Aβ such as phosphorylation and pyroglutamate modification. Erik Portelius of the University of Gothenburg in Sweden echoed Thal, pointing out that researchers will need to determine what characteristics—beyond size—distinguish neuroactive from benign Aβ aggregates. He added that if detectable in cerebrospinal fluid, these small oligomers would likely serve as very early markers of AD.

How might these findings inform therapeutic efforts to target Aβ? Selkoe said that targeting either small oligomers or their more abundant reservoirs of large oligomers could hold promise of slowing neuronal damage. He added that perhaps antibodies raised specifically against small oligomers dissociated from larger ones might directly target the most troublesome aggregates.

Lannfelt and colleagues have targeted the larger oligomers, also dubbed “protofibrils,” with a monoclonal antibody, BAN2401, that is now in Phase 2 clinical trials for AD. Other antibodies raised to Aβ epitopes also recognize oligomers and larger fibrils (see May 2015 news). 

“The hot debate at the moment is which of the current antibodies (and other compounds targeting Aβ oligomers) in clinical development have the best chance of either neutralizing the toxic effects, binding and activating microglial clearance pathways, or simply shifting the equilibrium between the different Aβ conformers,” commented Colin Masters of the University of Melbourne in Australia. Selkoe said his lab seeks to address that question by comparing the size of oligomers engaged by different therapeutic antibodies.—Jessica Shugart


  1. The experience from genetic findings in the early 1990s strongly point to Aβ as the culprit in Alzheimer’s disease. However, we still do not understand how Aβ confers cognitive dysfunctions and neuronal atrophy. For several years we have witnessed an increased interest in soluble Aβ oligomers as being the important pathogenic form of Aβ (Liu et al., 2015; Esparza et al., 2016). This paper by Yang et al. demonstrates that the predominant forms of soluble Aβ in human Alzheimer brain are higher molecular weight (HMW) oligomers. HMW oligomers eluted in the void volume of a size exclusion chromatography Superdex 75 column. The investigators then incubate these species in an alkaline buffer for 48 hours at 37°C, which leads to degradation into lower molecular weight (LMW) oligomers. The authors claim that toxicity of Aβ is mainly conferred by the small oligomers and not by the HMW oligomers, as demonstrated by effects by in vitro methods such as hippocampal LTP, activation of microglia and effect on neuronal levels of β2-adrenergic receptors.

    We have made a similar observation that the major soluble Aβ species in Alzheimer brain were HMW oligomers, but used ultracentrifugation. However, toxicity between synthetic HMW and LMW oligomers were approximately similar (Sehlin et al., 2012). By using synthetic Aβ we had previously found a fraction eluting in the void of a Superdex 75 column. This pool of oligomeric Aβ, of different sizes, we named protofibrils (Nilsberth et al., 2001). This name had been used previously (Walsh et al., 1997), and we found it appropriate to use the nomenclature of previous authors. The propensity to form protofibrils was greatly enhanced by the Arctic APP (E693G) mutation inside Aβ leading to early onset Alzheimer’s disease. We later developed an antibody against Aβ protofibrils, mAβ158 (Englund et al., 2007). The humanized version of the antibody, BAN2401, is now in a large clinical trial against Alzheimer’s disease, run by Eisai and BioArctic (Lannfelt et al., 2015). We have also used this antibody to visualize Aβ protofibrils in two AβPP-transgenic mouse models with PET (Sehlin et al., 2016), clearly demonstrating that these Aβ species are formed in vivo and are accessible to antibodies penetrating the brain.

    It is very difficult to study what is actually ongoing in the brains of Alzheimer patients. All our efforts and methods are indirect, as we at present are unable to monitor the biochemical processes that are taking place in vivo in the human brain. Thus, it is not certain that the in vitro studies used in the present paper mimic the disease processes in Alzheimer brains. Furthermore, it is unclear why the experiment was performed where the HMW material eluting in the void volume of the SEC column was subjected to degradation in alkaline conditions during 48 hours at 37°C. Is this something that reflects processes in the human Alzheimer brain? During 48 hours at 37°C many things can happen, including protease activity changing the nature of the oligomers.

    Solanezumab, a monoclonal antibody targeting the monomer of Aβ, recently failed to show clinical efficacy (see Dec 2016 conference news). However, aducanumab, an antibody with a very different Aβ-binding profile than solaneuzumab, has demonstrated a clinical effect on cognition and also reduction of Aβ plaque load as measured by amyloid PET (Sevigny et al., 2016). In contrast to solanezumab, the Aβ-binding profile of aducanumab is shifted to the “right,” i. e., toward HMW Aβ oligomers and fibrils. Sevigny et al. demonstrated that aducanumab mainly binds to Aβ fibrils and oligomers, with no monomer binding. If, as the authors suggest, the toxic effect of Aβ is caused by LMW oligomers in vivo, targeting HMW oligomers will still be beneficial since it will remove the proposed precursor of the LMW oligomers.

    With this in mind it is uncertain if the in vitro experiments in Yang et al. reflect the biochemical processes of Aβ aggregation in the brains of patients with Alzheimer’s disease. However, it is very important to understand that our scientific efforts are hampered by our lack of basic knowledge of the in vivo pathogenic processes in the human Alzheimer brain. Our conclusion is that our knowledge at present is limited, but there are certain evidences that soluble aggregated forms of Aβ are a main player in toxicity and confer neurodegeneration in Alzheimer’s disease. In previous papers some of these authors have strongly claimed that the Aβ dimer is the main cause of neurotoxicity (Shankar et al., 2008). Our position is that we do not fully understand the nature of what is killing neurons in Alzheimer’s disease, although soluble aggregated forms of Aβ are a prime suspect.

    Pär Gellerfors of BioArctic are also co-authors of this comment.


    . Sensitive ELISA detection of amyloid-beta protofibrils in biological samples. J Neurochem. 2007 Oct;103(1):334-45. PubMed.

    . Soluble Amyloid-beta Aggregates from Human Alzheimer's Disease Brains. Sci Rep. 2016 Dec 5;6:38187. PubMed.

    . Perspectives on future Alzheimer therapies: amyloid-β protofibrils - a new target for immunotherapy with BAN2401 in Alzheimer's disease. Alzheimers Res Ther. 2014;6(2):16. Epub 2014 Mar 24 PubMed.

    . Quaternary Structure Defines a Large Class of Amyloid-β Oligomers Neutralized by Sequestration. Cell Rep. 2015 Jun 23;11(11):1760-71. Epub 2015 Jun 4 PubMed.

    . The 'Arctic' APP mutation (E693G) causes Alzheimer's disease by enhanced Abeta protofibril formation. Nat Neurosci. 2001 Sep;4(9):887-93. PubMed.

    . Large aggregates are the major soluble Aβ species in AD brain fractionated with density gradient ultracentrifugation. PLoS One. 2012;7(2):e32014. PubMed.

    . Antibody-based PET imaging of amyloid beta in mouse models of Alzheimer's disease. Nat Commun. 2016 Feb 19;7:10759. PubMed.

    . The antibody aducanumab reduces Aβ plaques in Alzheimer's disease. Nature. 2016 Aug 31;537(7618):50-6. PubMed.

    . Amyloid-beta protein dimers isolated directly from Alzheimer's brains impair synaptic plasticity and memory. Nat Med. 2008 Aug;14(8):837-42. PubMed.

    . Amyloid beta-protein fibrillogenesis. Detection of a protofibrillar intermediate. J Biol Chem. 1997 Aug 29;272(35):22364-72. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Aβ Oligomers Purified from Human Brain
  2. Paper Alert: Patient Aβ Dimers Impair Plasticity, Memory
  3. San Diego: Oligomers Live Up to Bad Reputation, Part 1
  4. Spine Shrinkers: Aβ Oligomers Caught in the Act
  5. Shape of a Hug: How the Embrace of a Therapeutic Aβ Antibody Really Matters

Antibody Citations

  1. Aβ 1-42 (266)

Therapeutics Citations

  1. BAN2401

Paper Citations

  1. . Amyloid beta peptide-(1-42) induces internalization and degradation of beta2 adrenergic receptors in prefrontal cortical neurons. J Biol Chem. 2011 Sep 9;286(36):31852-63. PubMed.
  2. . Environmental novelty activates β2-adrenergic signaling to prevent the impairment of hippocampal LTP by Aβ oligomers. Neuron. 2013 Mar 6;77(5):929-41. PubMed.
  3. . Environmental Enrichment Potently Prevents Microglia-Mediated Neuroinflammation by Human Amyloid β-Protein Oligomers. J Neurosci. 2016 Aug 31;36(35):9041-56. PubMed.
  4. . High-molecular weight Aβ-oligomers and protofibrils are the predominant Aβ-species in the native soluble protein fraction of the AD brain. J Cell Mol Med. 2011 Mar 21; PubMed.
  5. . Large aggregates are the major soluble Aβ species in AD brain fractionated with density gradient ultracentrifugation. PLoS One. 2012;7(2):e32014. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Large Soluble Oligomers of Amyloid β-Protein from Alzheimer Brain Are Far Less Neuroactive Than the Smaller Oligomers to Which They Dissociate. J Neurosci. 2017 Jan 4;37(1):152-163. PubMed.