Of Four Aβ Antibodies, Only Aducanumab Stems Tide of Toxic Oligomers
A slew of anti-Aβ antibodies have been developed for clinical trials, in hopes that they will spur the immune system to remove amyloid-β from the brain. But might they also shift the dynamic of Aβ aggregation? According to a paper published today in Nature Structural and Molecular Biology, four monoclonal antibodies against different epitopes of the peptide interfere with dramatically different aspects of this process. Employing in vitro kinetic and biochemical approaches, researchers led by Sara Linse and Oskar Hansson, both at Lund University in Sweden, reported that while solanezumab disrupts the initial formation of fibrils, both bapineuzumab and gantenerumab slow their elongation. Aducanumab stemmed the formation of oligomers, a secondary nucleation process fueled by interactions between monomers and fibrils. It may work by coating the surface of fibrils, effectively removing them as a source of oligomerization. Only these four antibodies were tested. Biogen has filed for FDA approval for aducanumab for the treatment of Alzheimer’s disease.
- In vitro at least, investigational Aβ antibodies interfere with different phases of Aβ fibrillization.
- Solanezumab inhibits primary nucleation; bapineuzumab and gantenerumab slow elongation; aducanumab thwarts secondary nucleation of oligomers.
- Aducanumab may completely coat fibril surfaces, while other antibodies leave gaps.
“This paper really makes the point that these monoclonal antibodies are not doing the same thing—not by a long shot,” said Eric Siemers of Siemers Integration, LLC. While working at Lilly, Siemers oversaw clinical development of solanezumab. While it is generally accepted that oligomers are the most toxic of Aβ species, only results of clinical trials can tell whether antibodies that reduce them also prove to be the most efficacious, Siemers added.
Indeed, while plaques packed with Aβ fibrils are the obvious pathological hallmark of AD, studies have increasingly pointed to those ephemeral Aβ species—low-molecular-weight oligomers—as inflicting the most damage to neurons (Benilova et al., 2012; Reiss et al., 2018). Oligomers are in constant flux, complicating efforts to detect them or to gauge the effectiveness of monoclonal antibodies at vanquishing them.
Oligomers do not primarily form when monomers bump into one another; rather, scientists believe most are born of secondary nucleation (Dec 2009 news). This happens when monomers interact with the surface of Aβ fibrils, an encounter that not only makes monomers more likely to meet, but also goads them into oligomerizing.
Previously, Linse and colleagues had developed an in vitro assay system to study Aβ aggregation kinetics (May 2013 news). They found that each phase of the Aβ aggregation process—including the primary nucleation of monomers into fibrils, the elongation of fibrils, and the secondary nucleation of monomers into oligomers at the fibril surface—is marked by a unique kinetic signature, with its own rate constant (Cohen et al., 2012; Arosio et al., 2015; Dear et al., 2020).
Do monoclonal antibodies perturb different parts of this process? And do any of them throw a wrench into secondary nucleation? Linse and Hansson set out to test this. They asked various companies for samples of their antibodies. Biogen offered a mouse version of aducanumab, and also mouse versions of solanezumab, bapineuzumab, and gantenerumab that Biogen had been studying. Hansson told Alzforum that he tried hard to get other antibodies, including BAN2401, as well, but to no avail.
Of the four antibodies Linse and Hansson were able to obtain, each binds a different spot on the Aβ peptide, which ultimately dictates which species of Aβ it prefers (May 2015 news). Solanezumab has a strong penchant for Aβ monomers, because monomers display its epitope—residues 16-26—while in aggregates this epitope is buried. Gantenerumab binds aggregated forms of Aβ, including fibrils. Although its epitope, residues 3-11 and 18-27, is hidden within most fibrils, it may be exposed at their ends. Bapineuzumab and aducanumab latch on to amino acids 1-5 and 3-7, respectively, part of the peptide’s freewheeling N-terminal region, which has little secondary structure. While both epitopes are exposed in all species of Aβ, both antibodies bind tightest to aggregated forms of the peptide.
To test how the antibodies would shift the dynamics of Aβ fibrillization, the researchers added increasing concentrations of each to a solution of Aβ monomers, then turned up the temperature from zero to 37 degrees to initiate the aggregation process. In some reactions, the researchers added preformed aggregate seeds to bypass the primary nucleation step and focus on elongation and secondary nucleation. Using thioflavin T fluorescence to track aggregation over time, the researchers compared how each antibody tweaked the kinetics of each phase of aggregation.
In a nutshell, they found that solanezumab primarily disrupted primary nucleation, while gantenerumab and bapineuzumab nipped elongation of fibrils in the bud. Aducanumab inhibited secondary nucleation (see model below). At the lowest concentration used, 0.25 nM, aducanumab cut secondary nucleation by nearly half; at 100 nM, threefold. The scientists obtained similar results when they carried out the fibrillization in cerebrospinal fluid to mimic a bit more closely what might happen in the brain.
Assembly Line. In a solution assay, solanezumab (m266) preferentially blocked primary nucleation; bapineuzumab (3D6) and gantenerumab curtailed fibril elongation; aducanumab reduced secondary nucleation. [Courtesy of Linse et al., Nature Structural and Molecular Biology, 2020.]
Did disrupting secondary nucleation translate into a meaningful drop in Aβ oligomerization? To find out, the researchers attempted to measure oligomers biochemically—no small task. They used size-exclusion chromatography (SEC) followed by matrix-assisted laser desorption ionization (MALDI) mass spectrometry to detect free oligomers at the halfway point in their aggregation assay. With this method, they found that aducanumab reduced the concentration of these oligomers by more than 80 percent compared to an isotype control antibody. Bapineuzumab took oligomers down by about a third, while gantenerumab had no effect. In the presence of solanezumab, which latches onto monomers, aggregation never took off in the first place in this in vitro test, which means the researchers were unable to measure any secondary nucleation effects. This might be different in the brain of a person who has already accumulated plaques.
How did the antibodies’ interactions with Aβ relate to their mechanisms of action? To investigate, the researchers examined how each binds monomers and fibrils using a technique called microfluidic diffusional sizing. It quantifies interactions between macromolecules in real time within a flowing solution, rather than trying to capture inherently dynamic species with biochemical techniques (Arosio et al., 2016). Oligomers were too fickle to measure in this assay.
In agreement with previous studies, the scientists found that aducanumab bound fibrils four orders of magnitude more tightly than it did monomers. Notably, the diffusion data revealed one aducanumab molecule for every 4.5 Aβ42 monomers. This stoichiometry jibes with the size of the antigen-binding region of aducanumab, which spans the length of about five monomer rungs within a fibril. Together, the data paint a picture of aducanumab coating the entire surface of an Aβ fibril, explaining its rebuff of monomers from the fibril surface.
Both gantenerumab and bapineuzumab also preferred fibrils over monomers in this assay, but the stoichiometry indicates that these antibodies were spaced farther apart. Gantenerumab bound once every 44 monomers, while bapineuzumab bound once every 40. This is consistent with kinetics suggesting that these antibodies slowed fibril elongation more than secondary nucleation. Finally, solanezumab had a high affinity for monomers and did not bind fibrils at all, consistent with its inhibition of primary nucleation.
Finally, the researchers compared the kinetic effects of aducanumab with that of Brichos—a molecular chaperone that powerfully blocks secondary nucleation (Cohen et al., 2015). Though weaker than Brichos, aducanumab changed the kinetics of the aggregation in the same way, while the other antibodies did not (see image below).
Predictive Pentagrams? Brichos (gray), aducanumab (green), solanezumab (purple), bapineuzumab (blue), and gantenerumab (red) shift nucleation (pKn), elongation (pK+), and secondary nucleation (pK2) kinetic constants, as well as the binding constants for monomers (pKDm) and fibrils (pKDf). Colored slices depict the effect size, from zero at the center to maximal at the pentagram tips. Aducanumab works similarly to Brichos, affecting secondary nucleation and fibril binding. [Courtesy of Linse et al., Nature Structural and Molecular Biology, 2020.]
Overall, the findings suggest that while all four monoclonals block Aβ aggregation at the macroscopic level, they do so by meddling with different parts of the process. The finding that aducanumab derails oligomer secondary nucleation bodes well for potentially dampening amyloid toxicity, Linse said. However, she also noted that aducanumab was a relatively weak blocker of secondary nucleation compared to Brichos. Some of the authors co-founded Wren Therapeutics, which is now using kinetic assays to screen for small molecules that more powerfully and specifically derail secondary nucleation of Aβ, unlike Brichos.
This work did not study a host of other antibodies, previously or currently in clinical development, that might perturb Aβ aggregation, including crenezumab, donanemab, and BAN2401, which was developed specifically to target the most toxic protofibrils (Apr 2011 conference news).
“These findings suggest that aducanumab … should, as an early intervention or prophylaxis, result in cognitive benefit if the amyloid hypothesis holds water,” commented Luke Miles of St. Vincent’s Institute of Medical Research in Fitzroy, Australia. “If realized, these benefits might well be enhanced in combination with a specific inhibitor of primary oligomer nucleation, such as solanezumab,” he added.—Jessica Shugart
- Make or Break—Equation of Everything Fibrillar?
- Aβ Fibrils Drive Oligomer Formation, New Model Suggests
- Shape of a Hug: How the Embrace of a Therapeutic Aβ Antibody Really Matters
- Barcelona: Antibody to Sweep Up Aβ Protofibrils in Human Brain
- Benilova I, Karran E, De Strooper B. The toxic Aβ oligomer and Alzheimer's disease: an emperor in need of clothes. Nat Neurosci. 2012 Jan 29;15(3):349-57. PubMed.
- Reiss AB, Arain HA, Stecker MM, Siegart NM, Kasselman LJ. Amyloid toxicity in Alzheimer's disease. Rev Neurosci. 2018 Aug 28;29(6):613-627. PubMed.
- Cohen SI, Vendruscolo M, Dobson CM, Knowles TP. From macroscopic measurements to microscopic mechanisms of protein aggregation. J Mol Biol. 2012 Aug 10;421(2-3):160-71. Epub 2012 Mar 8 PubMed.
- Arosio P, Knowles TP, Linse S. On the lag phase in amyloid fibril formation. Phys Chem Chem Phys. 2015 Mar 28;17(12):7606-18. PubMed.
- Dear AJ, Meisl G, Michaels TC, Zimmermann MR, Linse S, Knowles TP. The catalytic nature of protein aggregation. J Chem Phys. 2020 Jan 31;152(4):045101. PubMed.
- Arosio P, Müller T, Rajah L, Yates EV, Aprile FA, Zhang Y, Cohen SI, White DA, Herling TW, De Genst EJ, Linse S, Vendruscolo M, Dobson CM, Knowles TP. Microfluidic Diffusion Analysis of the Sizes and Interactions of Proteins under Native Solution Conditions. ACS Nano. 2016 Jan 26;10(1):333-41. Epub 2015 Dec 23 PubMed.
- Cohen SI, Arosio P, Presto J, Kurudenkandy FR, Biverstål H, Dolfe L, Dunning C, Yang X, Frohm B, Vendruscolo M, Johansson J, Dobson CM, Fisahn A, Knowles TP, Linse S. A molecular chaperone breaks the catalytic cycle that generates toxic Aβ oligomers. Nat Struct Mol Biol. 2015 Mar;22(3):207-13. Epub 2015 Feb 16 PubMed.
No Available Further Reading
- Linse S, Scheidt T, Bernfur K, Vendruscolo M, Dobson CM, Cohen SI, Sileikis E, Lundqvist M, Qian F, O'Malley T, Bussiere T, Weinreb PH, Xu CK, Meisl G, Devenish SR, Knowles TP, Hansson O. Kinetic fingerprints differentiate the mechanisms of action of anti-Aβ antibodies. Nat Struct Mol Biol. 2020 Dec;27(12):1125-1133. Epub 2020 Sep 28 PubMed.
To make an annotation you must Login or Register.
The Florey Institute of Neuroscience
Linse et al. present a kinetic study of the influence of four amyloid-directed antibodies, developed to treat Alzheimer’s disease, on aspects of the complex process of amyloid-β aggregation. This is an advancement on our own efforts and those of others to develop a detailed understanding of the mechanisms of action of individual antibodies targeting Aβ in order to aid interpretation of differential outcomes from clinical trials (Miles et al., 2013; Watt et al., 2014; Crespi et al., 2015). At the heart of the matter at this time is the question of whether you can have antibodies targeting clearance or neutralization of Aβ that engage their target Aβ yet leave a pool of disease-causing amyloid, such as toxic soluble oligomers, untouched and in play. And if so, then how can we overcome this?
The authors report data on mouse versions of four clinical antibodies, namely aducanumab, bapineuzumab, solanezumab and gantenerumab. The side-by-side comparison of these therapeutics adds weight to this study because kinetic studies of amyloid are highly susceptible to misinterpretation owing to the pleiomorphic and temperamental nature of the ligand.
We have worked with biosimilar adu, bapi, and sola, and the authors’ general description of these antibodies is consistent with my own observations (sola, strong binder of monomers only; bapi, monomers and oligomers/fibrils; adu, oligomers/fibrils). Linse et al. show in an elegant breakdown of the complex amyloid aggregation pathway that sola readily binds monomeric Aβ and acts almost exclusively to inhibit primary aggregation of monomers. According to the study, both bapi and gant principally act to reduce fibril extension without significantly modifying fibril catalyzed (secondary) nucleation of Aβ aggregation (oligomer production). Aducanumab, which evades monomeric Aβ and binds oligomeric Aβ directly (potentially for clearance/neutralization) also significantly interferes with secondary nucleation of amyloid aggregation by interaction with fibrils. This is an important point because as noted in the papers’ introduction, above a certain level of fibrils, aggregation catalyzed at the fibril surface becomes the dominant driver of amyloid oligomer formation.
The authors go on to show that of the antibodies studied, only the effects of aducanumab are neutralized by competition with a known secondary nucleation inhibitor.
These findings suggest that monomer binding antibodies such as sola that make it into the brain without being inundated by Aβ in the periphery might still leave oligomeric assembly (toxicity) unaffected and where brain amyloid burden as measured by PET is stabilized but not cleared (bapi) that neuronal damage could continue unabated despite reduced brain amyloid burden. These findings also suggests that aducanumab, which is capable of robust amyloid clearance as measured by PET, acts as an inhibitor of amyloid oligomer nucleation, and should, as an early intervention or prophylaxis, result in cognitive benefit if the amyloid hypothesis holds water. These benefits if realized might well be enhanced in combination with a specific inhibitor of primary oligomer nucleation such as solanezumab.
Miles LA, Crespi GA, Doughty L, Parker MW. Bapineuzumab captures the N-terminus of the Alzheimer's disease amyloid-beta peptide in a helical conformation. Sci Rep. 2013 Feb 18;3:1302. PubMed.
Watt AD, Crespi GA, Down RA, Ascher DB, Gunn A, Perez KA, McLean CA, Villemagne VL, Parker MW, Barnham KJ, Miles LA. Do current therapeutic anti-Aβ antibodies for Alzheimer's disease engage the target?. Acta Neuropathol. 2014 Jun;127(6):803-10. Epub 2014 May 7 PubMed.
Crespi GA, Hermans SJ, Parker MW, Miles LA. Molecular basis for mid-region amyloid-β capture by leading Alzheimer's disease immunotherapies. Sci Rep. 2015 Apr 16;5:9649. PubMed.
Massachusetts General Hospital
Aducanumab was first developed by Roger Nitsch and Christoph Hock of Neurimmune when they started searching for naturally occurring autoantibodies to Aβ oligomers from memory B-cells of super agers. Roger is generous in his presentations to point out that the study that inspired the Neurimmune approach was carried out by the late Rob Moir, who passed away at 58 years old in December 2019 after a short battle with glioblastoma. Rob reported auto-antibodies that recognized cross-linked Aβ oligomeric species (CAPS) in human serum, and showed that levels of those antibodies correlated with decreased risk for Alzheimer's disease.
The final two sentences in his paper were: "We also observed that AD plasma contained lower levels of anti-CAPS antibodies compared to non-demented control subjects and that immunoreactivity to CAPS correlated with AAO of the disease. These findings may be useful for diagnosis and facilitating future designs of reagents for Aβ vaccination and antibody perfusion therapies aimed at treating and preventing AD."
Thank you, Rob. Rest in peace.
Moir RD, Tseitlin KA, Soscia S, Hyman BT, Irizarry MC, Tanzi RE. Autoantibodies to redox-modified oligomeric Abeta are attenuated in the plasma of Alzheimer's disease patients. J Biol Chem. 2005 Apr 29;280(17):17458-63. PubMed.
This paper provides useful contributions to amyloid-β antibody characterization with respect to their isoform kinetics, potency, and selectivity prior to clinical testing. Two small additional comments on why Aβ monomer-selective antibodies like aducanumab seem to be preferential clinical candidates.
It is wonderful to see careful mechanistic characterization of these important phenomena. I believe ProMIS Neurosciences has been making a similar case regarding its PMN310 with higher selectivity and affinity for the correct target.
Disclosure: I have made a personal investment in ProMIS Neurosciences (ARFXF).View all comments by Martin Page
The authors claim, based on this modeling, that aducanumab selectively reduces the secondary nucleation rate, solanezumab selectively inhibits primary nucleation, and bapineuzumab and gantenerumab act by reducing elongation of fibrils. Furthermore, the effect by aducanumab is claimed to be caused by the antibody’s interaction with Aβ species involved in secondary nucleation along the surface of fibrils, leading to reduction of oligomers. This should, according to the authors, explain the clinical efficacy of aducanumab versus the other three antibodies. Moreover, the authors intriguingly suggest that aducanumab is the only antibody of the four which has the same kinetic “fingerprint” as the chaperone Brichos. Brichos has recently been proposed to act as a chaperone to prevent amyloid toxicity (Cohen et al., 2015; Nerelius et al., 2008). The definition of a molecular chaperone is to assist the conformational folding or unfolding and the assembly or disassembly of other macromolecular structures to form the active three-dimensional structure.
We believe that the molecular details in the complex Aβ aggregation processes are still not completely understood, and different models have been proposed for Aβ aggregation. It remains to be seen if aducanumab holds these unique properties. An obvious question is why an antibody against Aβ should have chaperone activity? The main function of anti-Aβ antibodies, in our view, is to remove Aβ by microglial phagocytosis, not dissolve and thereby provide more Aβ substrate to feed into the aggregation pathway.
We have compared aducanumab and BAN2401 with inhibition-ELISA and surface plasmon resonance (SPR). We aimed to describe the binding pattern of the two antibodies to different species of Aβ. We found that BAN2401 had a stronger binding to all forms of Aβ, including monomers, oligomers, small protofibrils, large protofibrils, and fibrils, than did aducanumab. Furthermore, BAN2401 bound strongest to small and large protofibrils, whereas aducanumab bound strongest to fibrils. In presentations at AAIC and CTAD last year we reported that BAN2401 bound stronger to small oligomers than did aducanumab. It remains to be seen if the different assays used by us and others will explain the outcome of the ongoing Phase 3 clinical trials for aducanumab, gantenerumab, and BAN2401.
Overall, it might be premature to claim that aducanumab uniquely and dramatically reduces the flux of Aβ oligomers based only on an in vitro observation and modeling exercises. The findings by Linse et al. provide interesting new hypotheses for how different anti-Aβ antibodies might act, but ultimate translatability to the clinic and the impact on Aβ remains unclear.
Cohen SI, Arosio P, Presto J, Kurudenkandy FR, Biverstål H, Dolfe L, Dunning C, Yang X, Frohm B, Vendruscolo M, Johansson J, Dobson CM, Fisahn A, Knowles TP, Linse S. A molecular chaperone breaks the catalytic cycle that generates toxic Aβ oligomers. Nat Struct Mol Biol. 2015 Mar;22(3):207-13. Epub 2015 Feb 16 PubMed.
Nerelius C, Martin E, Peng S, Gustafsson M, Nordling K, Weaver T, Johansson J. Mutations linked to interstitial lung disease can abrogate anti-amyloid function of prosurfactant protein C. Biochem J. 2008 Dec 1;416(2):201-9. PubMed.
Linse et al. have published an excellent and elegant biophysical study characterizing the targets and therapeutic mechanisms of four clinical Aβ antibodies. The results provide yet more evidence that disease-modifying activity in AD requires targeting of toxic oligomers, directly or indirectly.
The article demonstrates and quantifies aducanumab blockade of oligomer production by secondary nucleation in a reductionist in vitro system, using synthetic Aβ peptide. But in view of the well-established binding of aducanumab to oligomers in vitro and in vivo, these results do not rule out the possibility that aducanumab could also possess therapeutic activity by direct neutralization or clearance of toxic oligomers in Alzheimer brain, as suggested by Sevigny et al. in their publication of the PRIME trial (2016).
Several lines of evidence in fact indicate that direct targeting of oligomers, regardless of their origin (primary or secondary nucleation) is associated with efficacy. For example, BAN2401, which reacts with a different epitope on oligomers (and also interacts with fibrils, but to a lesser degree than aducanumab), has yielded positive results in Phase 2 trials. Moreover, approaches with greater oligomer selectivity have shown efficacy in vivo such as the PRI-002 anti-oligomer peptide (Schemmert et al., 2019) and ProMIS Neurosciences’ oligomer-selective antibodies (Silverman et al., 2018; Gibbs et al., 2019. Full disclosure: I am a co-founder and CSO of ProMIS).
The humanized ProMIS Neurosciences PMN310 antibody selectively binds and neutralizes toxic oligomers without interacting with Aβ monomers or insoluble fibrils. This antibody profile should promote improved safety and efficacy by focusing the dose on the toxic oligomer species of Aβ and avoiding interaction with plaque and vascular deposits which has been associated with the development of ARIA (Sperling et al., 2012; Carlson et al., 2016).
Sevigny J, Chiao P, Bussière T, Weinreb PH, Williams L, Maier M, Dunstan R, Salloway S, Chen T, Ling Y, O'Gorman J, Qian F, Arastu M, Li M, Chollate S, Brennan MS, Quintero-Monzon O, Scannevin RH, Arnold HM, Engber T, Rhodes K, Ferrero J, Hang Y, Mikulskis A, Grimm J, Hock C, Nitsch RM, Sandrock A. The antibody aducanumab reduces Aβ plaques in Alzheimer's disease. Nature. 2016 Aug 31;537(7618):50-6. PubMed.
Silverman JM, Gibbs E, Peng X, Martens KM, Balducci C, Wang J, Yousefi M, Cowan CM, Lamour G, Louadi S, Ban Y, Robert J, Stukas S, Forloni G, Hsiung GR, Plotkin SS, Wellington CL, Cashman NR. A Rational Structured Epitope Defines a Distinct Subclass of Toxic Amyloid-beta Oligomers. ACS Chem Neurosci. 2018 Jul 18;9(7):1591-1606. Epub 2018 Apr 16 PubMed.
Gibbs E, Silverman JM, Zhao B, Peng X, Wang J, Wellington CL, Mackenzie IR, Plotkin SS, Kaplan JM, Cashman NR. A Rationally Designed Humanized Antibody Selective for Amyloid Beta Oligomers in Alzheimer's Disease. Sci Rep. 2019 Jul 8;9(1):9870. PubMed.
Schemmert S, Schartmann E, Zafiu C, Kass B, Hartwig S, Lehr S, Bannach O, Langen KJ, Shah NJ, Kutzsche J, Willuweit A, Willbold D. Aβ Oligomer Elimination Restores Cognition in Transgenic Alzheimer's Mice with Full-blown Pathology. Mol Neurobiol. 2018 Jul 12; PubMed.
Sperling R, Salloway S, Brooks DJ, Tampieri D, Barakos J, Fox NC, Raskind M, Sabbagh M, Honig LS, Porsteinsson AP, Lieberburg I, Arrighi HM, Morris KA, Lu Y, Liu E, Gregg KM, Brashear HR, Kinney GG, Black R, Grundman M. Amyloid-related imaging abnormalities in patients with Alzheimer's disease treated with bapineuzumab: a retrospective analysis. Lancet Neurol. 2012 Mar;11(3):241-9. PubMed.
Carlson C, Siemers E, Hake A, Case M, Hayduk R, Suhy J, Oh J, Barakos J. Amyloid-related imaging abnormalities from trials of solanezumab for Alzheimer's disease. Alzheimers Dement (Amst). 2016;2:75-85. Epub 2016 Mar 2 PubMed.
Make a Comment
To make a comment you must login or register.