How short peptides can disassemble ultra-stable tau fibrils extracted from Alzheimer’s disease brain by a strain-relief mechanism

How short peptides can disassemble ultra-stable tau fibrils extracted from Alzheimer’s disease brain by a strain-relief mechanism | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Biological Sciences - Article How short peptides can disassemble ultra-stable tau fibrils extracted from Alzheimer’s disease brain by a strain-relief mechanism David Eisenberg, Ke Hou, Peng Ge, Michael Sawaya, Joshua Dolinsky, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4152095/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Jul, 2025 Read the published version in Nature → Version 1 posted You are reading this latest preprint version Abstract Reducing fibrous aggregates of protein tau is a possible strategy for halting progression of Alzheimer's dis-ease (AD). Previously we found that in vitro the D-peptide D-TLKIVWC disassembles tau fibrils from AD brains (AD-tau) into benign segments with no energy source present beyond ambient thermal agitation. This disassembly by a short peptide was unexpected, given that AD-tau is sufficiently stable to withstand disas-sembly in boiling SDS detergent. To consider D peptide-mediated disassembly as a potential therapeutic for AD, it is essential to understand the mechanism and energy source of the disassembly action. We find as-sembly of D-peptides into amyloid-like fibrils is essential for tau fibril disassembly. Cryo-EM and atomic force microscopy reveal that these D-peptide fibrils have a right-handed twist and embrace tau fibrils which have a left-handed twist. In binding to the AD-tau fibril, the oppositely twisted D-peptide fibril produces a strain, which is relieved by the disassembly of both fibrils. This strain-relief mechanism appears to operate in other examples of amyloid fibril disassembly and provides a new direction for the development of first-in-class therapeutics for amyloid diseases. Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy Biological sciences/Neuroscience/Diseases of the nervous system/Alzheimer's disease Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Tau pathology caused by the abnormal aggregation of tau is more strongly correlated with cognitive symptoms and severity in Alzheimer's disease (AD) than Aβ plaques 1–3 . Thus, disrupting tau aggregates emerges as a promising therapeutic strategy for AD 4,5 . To date, several types of disaggregators of tau fibrils have been reported and their disaggregation mechanisms are being sought. For example, the human Hsp70/DnaJ system can disaggregate tau fibrils in vitro in an ATP-dependent manner, but it generates small, seeding-competent species that accelerate the progression of disease 6 . Small molecules such as methylene blue 7 and EGCG, by binding to tau fibrils, apparently disrupt intermolecular interactions 8 . Our recent structural studies revealed EGCG stacks in the clefts formed at the junction of the two protofilaments in AD tau fibrils 9 . However, small molecules including methylene blue/LMTX and EGCG have not been proven to be effective drugs, perhaps because of limited bioavailability, promiscuous protein binding and low blood-brain barrier permeability 10 . A peptide-based disaggregator can offer advantages over small molecules including higher specificity and higher binding affinity 11 . Especially, compared to L-peptides, D-enantiomeric peptides are known to be less immunogenic, less protease-sensitive in vitro and more resistant to degradation in vivo 12 . A recent phase I clinical trial of a D-peptide that disassembles Aβ oligomers has proved to be safe and well tolerated 13 . Previously, we reported that the D-peptide D-TLKIVWC can disassemble tau fibrils extracted from AD brains (AD-tau), neutralizing their seeding ability and rescuing behavioral deficits in a mouse model of Alzheimer’s disease 14 . However, the underlying disassembly mechanism remains unknown, preventing further development of this type of drug candidate for AD. Here, we designed a series of peptides of sequence D-TLKIVWX varying only at the seventh residue, X. These D-peptides showed variable efficacy in disassembling AD-tau fibrils in vitro , with X = Ile as the best performer. From electron microscopy, we discovered that D-TLKIVWX peptides form amyloid-like fibrils themselves, and from atomic force microscopy we learned that these fibrils have a right-handed helical twist, in contrast to the left-handed helical twist of AD-tau. To learn the molecular interactions between the fibrils that participate in disassembly, we determined the cryo-EM structures of D-TLKIVWX protofilaments bound to tau fibrils of opposing twist. Combining our structural data, we propose a strain-relief mechanism for AD-tau fibril disassembly. D-TLKIVWX peptides disassemble AD-tau fibrils to non-seeding species In previous work, we found that D-TLKIVWC can disassemble tau fibrils, whereas its six-residue analog D-TLKIVW cannot 14 . Here we wondered whether the disassembling ability arises from the capacity of the cysteine to form a disulfide bond with tau residue Cys-322 located in the structured core of tau fibrils. The formation of a disulfide bond could alter the conformation of tau in fibrils, resulting in fibril breakdown 15 , 16 . To evaluate this hypothesis, we measured disassembly activity as a function of disulfide bond forming potential which we adjusted by including either glutathione (GSH) or glutathione disulfide (GSSG). As shown in Extended Data Fig. 1a, D-TLKIVWC exhibited equivalent capacity to disassemble recombinant tau K18+ (residues Gln244-Glu380 of 4R tau) fibrils across all tested conditions, indicating the formation of a disulfide bond is irrelevant to the disassembly of tau fibrils. To determine if the cysteine residue of D-TLKIVWC is essential in disassembling tau fibrils, we substituted the D-cysteine with various other D-amino acid residues, including anionic residues (aspartate (D) and glutamate (E)), cationic residues (arginine (R) and lysine (K)), polar residues (serine (S) and threonine (T)), hydrophobic residues (alanine (A), isoleucine (I), and valine (V)), as well as a β-sheet interrupter (proline (P)) 17 . Their performance in disassembling AD-tau fibrils after 48 hours of incubation was evaluated by dot blot and electron microscopy (TEM). The dot blot experiment was conducted with the conformational antibody GT38 18 , which specifically recognizes AD-tau fibrils and does not probe monomeric tau or D-peptide controls (Extended Data Fig. 1b). The D-peptide variants (D-TLKIVWX, X = A, S, D, I, V, R, K, E, T) showed varying efficacy in disassembling AD-tau fibrils depending on the type of residue in the seventh position (Fig. 1a-b). Specifically, hydrophobic residues (X = I, V and A) most significantly reduced the level of AD-tau fibrils, with efficacy decreasing in the following order: hydrophobic > polar (X = S and T) > cationic (X = R and K) > anionic (X = D and E). Lastly, the β-sheet interrupter (X = P) and the deletion of the seventh residue both showed no reduction in the level of AD-tau fibrils. This trend was consistently supported by TEM characterization, where the number of AD-tau fibrils (labeled by red arrows in Fig. 1c and Extended Data Fig. 1c) was quantified after 48 hours of disassembly (Extended Data Fig. 1d). Notably, we observed new fibril species (blue arrows in Fig. 1c and Extended Data Fig. 1c). Furthermore, we observed specificity of D-TLKIVWX in disassembling AD-tau fibrils as it cannot disrupt other amyloid fibrils such as α-syn fibrils or wild type hnRNPA2 fibrils (Extended Data Fig. 2). In summary, most tested residue types X in D-TLKIVWX demonstrate specific disassembly action against AD-tau fibrils, with hydrophobic residues, especially Ile, being the best. In addition to uncovering the mechanism of disassembly, it is important to ascertain whether disassembly produces pathologic products that seed the growth of additional fibrils 19 . Therefore, we systematically investigated the seeding capacity of the products of AD-tau fibrils after overnight incubation with variants of D-TLKIVWX in a HEK293T cell line stably expressing yellow fluorescent protein (YFP)-fused tau 20 . As illustrated in Fig. 1d, AD-tau fibrils alone can seed the aggregation of endogenous fluorescent tau, leading to the formation of bright puncta (indicated as white arrows) observable under fluorescence microscopy. In contrast, the overnight disassembly products of AD-tau fibrils treated with D-TLKIVWX gradually lost their seeding ability in a dose-dependent manner (Fig. 1d and Extended Data Fig. 3a). Further, automated image analysis of visible puncta revealed a dependence of the seeding capacity on the seventh amino acid residue within D-TLKIVWX (Extended Data Fig. 3b-f). The observed dependence is consistent with the trends observed in our in vitro disassembly assays (Fig. 1b and Extended Data Fig. 1c,d). Additionally, because of disassembly activity, D-TLKIVWX (X = I and S) shows a dose-dependent effect in reducing AD-tau toxicity in mouse Neuro 2A (N2a) cells (Fig. 1e and Extended Data Fig. 3g). Our results demonstrate that the products of AD tau fibrils disassembled by D-TLKIVWX are not seeding-competent and are non-toxic. Amyloid-like fibril formation of D-TLKIVWX is essential for tau disassembly To better understand the disassembly process of tau fibrils, we conducted a time-course dot blot and TEM experiment using D-TLKIVWI as a representative of the more efficient fibril disassemblers. Dot blot data (Extended Data Fig. 4a) confirmed that D-TLKIVWI gradually reduced the level of AD-tau fibrils as a function of time. The time-course TEM images in Fig. 2a and Extended Data Fig. 4b show AD-tau fibrils (red arrows) appearing to be covered by unknown species after one hour of incubation with D-TLKIVWI, and additional fibrillar structures (blue arrows) become increasingly evident at three- and six-hour timepoints. At 24 h, amorphous products (magenta arrows) appeared concomitant with the disappearance of AD-tau fibrils and reduction of the unidentified fibril species. The amorphous products appeared more dispersed in micrographs acquired at 48 h, and a western blot showed that the disassembly products of AD-tau fibrils consisted primarily of insoluble species; denatured pelleted material migrated as dimers and other multimers, not as monomeric tau (Extended Data Fig. 4c). Notably, in the TEM images of D-TLKIVWX (X = I, S, R, D, E, K, T, C, A and V)-treated AD-tau samples (Fig. 1c and Extended Data Fig. 1c), we also observed the emergence of new fibrils (labeled by blue arrows) accompanying with the disappearance of AD-tau fibrils. We identified these new species as D-TLKIVWX fibrils since D-TLKIVWX (X = I, S, R, D, E, K, T, C, A and V) exhibited aggregation activity in the same buffer (Extended Data Fig. 5a). In contrast, neither D-TLKIVW nor D-TLKIVWP exhibited aggregation nor disassembled AD-tau fibrils. These observations suggest that the ability of D-TLKIVWX to fibrilize aids the disassembly of AD-tau fibrils. To test the hypothesis that D-TLKIVWX form amyloid-like fibrils that disassemble AD-tau, we designed a negative control experiment by eliminating the ability of D-TLKIVWX to fibrilize. As shown in Fig. 2b and Extended Data Fig. 5b,c, 10 mM D-TLKIVWX (X = I, S and R) form well-defined fibrils when they are left undisturbed for three days at room temperature. The corresponding X-ray diffraction analysis (Fig. 2c and insets in Extended Data Fig. 5b,c) confirm these D-TLKIVWX (X = I, S and R) fibrils exhibit the characteristic features of amyloid fibrils with a strong 4.7-4.8 Å reflection corresponding to the distance between hydrogen-bonded β-strands, and a more diffuse 8-12 Å ring arising from the inter-sheet spacing 21 . As such, the aggregation activity of D-TLKIVWX might be prevented by eliminating hydrogen bonds between neighboring β-strands through N-methylation of peptide backbones 22 . Indeed, when we N-methylated the D-isoleucine of D-TLKIVWX (named as D-TLK(N-Me-I)VWX (X = I, S and R)), the peptides were unbale to form fibrils, as shown in Fig. 2d and Extended Data Fig. 5d,e. TEM and dot blot experiments further showed that these non-self-aggregating peptides D-TLK(N-Me-I)VWX (X = I, S and R) were unable to disassemble AD-tau fibrils (Fig. 2e-g and Extended Data Fig. 5f,g). Taken together, these experiments demonstrate the critical role of amyloid-like characteristics of D-TLKIVWX in disassembling AD-tau fibrils. D-TLKIVWX peptides form right-handed fibrils with conserved motifs We determined the structures of D-TLKIVWX amyloid-like fibrils, aiming to reveal features that facilitate disassembly of AD-tau fibrils. Generally, β-sheets formed by L-peptides adopt left-handed helical structures, while β-sheets formed by D-peptides adopt right-handed helical structures 23,24 . Here, atomic force microscopy (AFM) confirmed the twist is right-handed in all 18 polymorphs of D-TLKIVWI fibrils observed 25,26 , 27 (Fig. 3a and Extended Data Fig. 6a). Using cryo-EM, we were able to determine the structures of the predominant polymorphs of D-TLKIVWX (X = I, S and R) fibrils (indicated by red squares in Extended Data Fig. 6b-d) at 3.6 Å, 3.5 Å, 3.7 Å resolution, respectively (Fig. 3b-d). Data collection and refinement statistics are summarized in Supplementary Table 1. The D-TLKIVWX (X = I, S and R) fibrils are each composed of different numbers of protofilaments and these protofilaments associate in different patterns, but all the protofilaments share the same underlying structural motif known as a "steric zipper" 28 , a pair of β-sheets mated together by an interface of snugly fitting sidechains (Fig. 3b-d and Extended Data Fig. 7a,b). In addition, the steric zippers formed by D-TLKIVWX (X = I, S and R) all share the same symmetry pattern in which antiparallel β-sheets mate together by interfacing sidechains of Leu2, Ile4, and Trp6 (an example of “class 5” symmetry 29 ). As a result, the helical rise of D-TLKIVWX (X = I, S and R) amyloid-like fibrils is 9.56 Å, instead of the 4.80 Å spacing that is common among pathogenic amyloid fibrils. Notably, the sidechain of the seventh residue faces outward from the steric zipper interface in all cases. The identity of the seventh sidechain appears to affect the geometry of association between protofilaments but does not affect the symmetry of the protofilament itself. Thus, the ability to disassemble AD-tau fibrils seems linked more strongly to the steric zipper symmetry (which is conserved among all three functional D-peptides), rather than the pattern of association between zippers (which differs among the three D-peptides). Importantly, the structures suggest why the absence of a seventh residue in D-TLKIVW prohibits its fibril formation, and therefore presumably its inability to disassemble tau. Removal of the seventh residue would destabilize fibril formation by reducing the number of backbone hydrogen bonds (10 vs. 14) and increasing the distance between the negative charge of the C-terminal carboxylate and the compensating positive charge at the N-terminal amine of the adjacent strand (5.0 Å vs. 2.8 Å) (Extended Data Fig. 7c). D-TLKIVWX protofilaments grow along left-handed AD-tau fibrils To elucidate how D-TLKIVWX might interact with AD-tau fibrils to induce their disassembly, we cryogenically trapped complexes of D-TLKIVWX (X =I, S and R) with AD-tau fibrils at an intermediate 24-h time point with a lower ratio of D-TLKIVWX to tau (estimated 100:1) in comparison with the time-course experiment in Fig. 2a (500 :1). As a negative control, we also collected images of AD-tau fibrils in the absence of D-TLKIVWX. Helical reconstructions of the AD-tau complexed with each of the three D-peptides revealed the paired helical filament (PHF) tau polymorph 30 (Extended Data Fig. 8), and the atoms modeled into the PHF density showed no significant structural deviations from the negative control (Supplementary Table 2). However, the cryo-EM map of PHF complexed with D-TLKIVWX (X =I, S and R) revealed residual density near Val313-Thr319 of PHF, which was absent from our control (Fig. 4a-c and Extended Data Fig. 9a,b). We attribute the residual density to D-TLKIVWX for two reasons: (1) the shape of the residual density resembles one steric zipper unit of D-TLKIVWX fibrils (Fig. 4a-c and Fig. 3b-d); (2) 1 H- 15 N-HSQC NMR experiments indicated that D-TLKIVWX (X =I and S) interacted with 15 N, 13 C-labeled tau K18+ monomer 31 with chemical shift perturbation mapped to Val306-Lys311 11 and Val313-Thr319 (Fig. 4d and Extended Data Fig. 9c,d), consistent with the location of additional density next to tau PHF. Note that a rapid precipitation of D-TLKIVWX occurred upon mixing in our NMR experiments, corresponding to initial D-peptide fibril formation, but chemical shift changes of monomeric tau were observed over time when soluble fraction of D-peptides increased. Refinement of the 3D reconstruction of D-TLKIVWX (X =I, S and R) complexed with Tau PHF achieved overall resolutions of 3.1 Å, 3.1 Å and 3.5 Å, respectively. We note that the cryo-EM density map of D-TLKIVWI is stronger than that of D-TLKIVWS/R and that D-TLKIVWR is slightly further from the core of tau PHF (Fig. 4a-c and Extended Data Fig. 9b). This difference in occupancy and positioning may explain their unequal efficiency in disassembling AD-tau fibrils (Fig. 1b). Discussion Strain-relief of D-TLKIVWX fibrils drives disassembly of AD-tau fibrils Our observations lead us to propose the strain-relief mechanism of amyloid fibril disassembly illustrated in Fig. 5. As D-TLKIVWX binds to the Tau PHF, it stacks to form D-TLKIVWX amyloid-like protofilaments and these protofilaments are constrained by binding to adopt the left-handed helical twist of Tau PHF (Fig. 5a,b). Because D-TLKIVWX fibrils have an intrinsic right-handed twist (Fig. 3), the cryo-trapped structure of Figure 4a-c is metastable, and a strain develops. If not trapped by freezing, this metastable structure disassembles in hours. We propose that the strain produced by the metastable left-hand twist of D-TLKIVWX is relieved by the protofilament reversing its twist from a left- to right-handed helix (Fig. 5c). Because the D-peptide fibrils is bound to Tau PHF fibril, the reversal pulls against the Tau PHF structure. As a result, the tau residues that contact D-TLKIVWX are torn away from the native PHF contacts, thereby breaking backbone hydrogen bonds between tau molecules, and permitting solvent to invade the fibril core and further dissociate tau PHF (Fig. 5d). The disassembly products are non-seeding species (Fig. 1); they are tau-D-TLKIVWX complexes, not tau monomers (Extended Data Fig. 4c). This dynamic process is visually depicted in the Supplementary Video 1. Our proposed strain-relief mechanism is consistent with all our experimental findings. First, D-TLKIVWX disassembles tau fibrils to which its precursor, D-TLKIVW, was designed to bind, but does not disassemble other left-handed amyloid fibrils, such as α-synuclein and wildtype hnRNPA2 LCD fibrils. Second, other analogs of D-TLKIVWX, including D-TLKIVW, D-TLKIVWP, and D-TLK(N-Me-I)VWX (X = I, S, R) fail to disassemble AD-tau fibrils, presumably because they lack the ability to form fibrils themselves. Third, the variation in efficacy of disassembly AD-tau fibrils among D-TLKIVWX can be influenced by their binding strength with AD-tau fibrils. Fourth, the concurrent disappearance of AD-tau fibrils and the newly formed D-TLKIVWX fibrils, along with the emergence of amorphous products at later time points observed in the time-course EM images (Fig. 2a) strongly suggests that the AD-tau fibrils are disassembled. An additional observation that supports this “strain-relief” mechanism of disassembly is that L-TLKIVWX (X = C, I, S, R) all display inferior efficacy in disassembling AD-tau fibrils compared to their enantiomers D-TLKIVWX (Extended Data Fig. 10a,b). Despite exhibiting the same fibril-forming property as D-TLKIVWX 32 (Extended Data Fig. 10c) and nearly identical binding to tau monomers (Extended Data Fig. 10d,e), the L-TLKIVWI fibrils principally possess a left-handed twist, same as tau PHF. Therefore, the L-TLKIVWX would have less structural torsion to release when they bind and assemble along the axis of tau PHF in comparison with D-TLKIVWX, resulting in their inferior efficacy in disassembling AD-tau fibrils. The strain-relief mechanism may be a general theme of action of disruptors of amyloid fibrils. In previous work, we presented evidence that the polyphenolic compound EGCG disassembles AD-tau fibrils by stacking into a metastable amyloid-like fibril on the surface of AD-tau fibrils 9 . A subsequent change in which aromatic rings of EGCG curve into a more stable conformation could provide the energy to disassemble stable AD-tau fibrils. Thus, both disassembling actions of the very different compounds EGCG and D-TLKIVWX on AD-tau fibrils may be considered examples of strain-relief mechanisms. In summary, we find the disassembly of AD-tau fibrils is not exclusive to D-TLKIVWC, because D-TLKIVWX (X = A, S, D, I, V, R, K, E, T) also displays this property, but with unequal efficacy. We find that the amyloid-like, fibril-forming property of D-TLKIVWX contributes to the disassembly of AD-tau fibrils. Based on cryo-EM, AFM, NMR, and other results reported here, we propose that the disassembly of AD-tau fibrils is driven by the release of torsion in D-TLKIVWX protofilaments. It remains unknown whether D-TLKIVWX disassembles tau fibrils from other tauopathies 33 . The strain-relief mechanism of amyloid disassembly proposed here may explain how diverse small molecules can provide sufficient energy to disrupt extremely stable pathogenic amyloid fibrils. This mechanism may be applied to the design of a new generation of disaggregators for tau and other pathological amyloids. Methods Chemicals and Materials All the peptides were synthesized by GenScript and purified to ≥98%, as determined by mass spectrometry and HPLC (GenScript Corp, Piscataway, NJ). Recombinant protein expression and purification Unlabeled recombinant tau K18+ (residues Gln244-Glu380 of 4R tau) was expressed in a pNG2 vector in BL21-Gold E. coli cells grown in LB to an A 600 = 0.8. Cells were induced with 0.5 mM isopropyl 1-thio-β-D-galactopyranoside (IPTG) for 3 h at 37 °C and lysed by sonication in 20 mM MES buffer (pH 6.8) with 1 mM EDTA, 1 mM MgCl 2 , 1 mM dithiothreitol (DTT), and HALT protease inhibitor before the addition of NaCl (500 mM final concentration). The lysate was boiled for 20 min and then clarified by centrifugation at 15,000 rpm for 15 min and dialyzed to 20 mM MES buffer (pH 6.8) with 50 mM NaCl and 5 mM DTT. Dialyzed lysate was purified on a 5-mL HiTrap SP HP ion exchange column and eluted over a gradient of NaCl from 50 to 550 mM. Protein was further purified on a HiLoad 16/600 Superdex 75 pg column (GE Healthcare) in 10 mM Tris (pH 7.6) with 100 mM NaCl and 1 mM DTT and concentrated to 20-60 mg/mL by ultrafiltration using a 3-kDa cutoff filter (Millipore-Sigma, Burlington, MA). Isotopically labeled tau K18+ proteins for NMR were grown in M9 H 2 O media supplemented with 15 NH 4 Cl (and 13 C-glucose) as the sole nitrogen (and carbon) source. Protein expression was induced with 1 mM IPTG at 37 °C for 6 hours. The purification was same as for the unlabeled protein. The construct for overexpression of mCherry-hnRNPA2-LCD fusion protein was provided by Dr. Masato Kato of University of Texas, Southwestern. Protein overexpression and purification procedures followed the same protocol reported previously 34 . Extraction of tau fibrils from AD patient brains Frozen brain tissues were weighed, diced into small pieces, and resuspended in 10 mL/gram of sucrose buffer (800 mM NaCl, 10% sucrose, 10 mM Tris-HCl, pH 7.4, 0.1 mM EDTA, 1 mM DTT) supplemented with 1:100 (v/v) Halt protease inhibitor (Thermo Scientific). Resuspended tissue was homogenized using a Polytron homogenizer (Thomas Scientific) and centrifuged at 20,000 × g at 4 °C for 20 minutes. The crude supernatant was treated with N-lauroylsarcosinate (1% [w/v] final concentration) and shaken at room temperature (22 °C) for 1 h. The supernatant was then centrifuged at 100,000 × g for 1 h at 4 °C. The sarkosyl-insoluble pellet was resuspended in washing buffer (10 mM Tris–HCl, pH 7.4, 800 mM NaCl, 5 mM EDTA, 1 mM EGTA, 1 mM DTT, 10% sucrose) and centrifuged at 20,100 x g for 30 mins at 4 °C. After centrifugation, the supernatant was further centrifuged at 100,000 × g (Beckman Coulter, Optima MAX-XP) for 1 h at 4 °C. Finally, the purified sarkosyl-insoluble pellet was resuspended in 250 μl 20 mM Tris-HCl, pH 7.4, 100 mM NaCl and stored at -80 °C. Thioflavin T (ThT) assay Kinetic fluorescence data were collected in a microplate reader (FLUOstar Omega, BMG Labtech) at 37 ˚C with double orbital shaking at 700 rpm. Fluorescence measurements were recorded every 10 mins with excitation and emission wavelengths of 440 and 480 nm. All samples were added in triplicate and experiments were repeated at least twice. Dot blot assay Purified AD-tau fibrils from brain extract were incubated with 500 μM L/D-TLKIVW, D-TLKIVWX (X = C, A, S, D, I, V, R, K, E, P, T), D-TLK(N-Me-I)VWX (X =I, S and R), L-TLKIVWX (X = C, I, S and R) at 37 ℃ for 48 h, respectively. 2.5 μL of samples were added on nitrocellulose membrane (0.2 µm, Bio-Rad). The membrane was blocked by 5% (w/v) nonfat dry milk in TBS-T (T = 0.1% (v/v) Tween-20) at room temperature for 1 hr. After blocking, the membrane was incubated with GT38 antibody obtained from Virginia Lee’s lab (1:1000) in 5% (w/v) milk in TBS-T at 4 ℃ overnight. Then, the membrane was washed in TBS-T three times for 5 minutes each and incubated with goat anti-mouse IgG HRP (1:5000, cat# AB205719, Abcam) in TBS-T for 1 h at room temperature. The membrane was washed three more times, and the signal was developed with Pierce TM ECL western blotting substrate (170-5061, BioRad). Negative stain transmission electron microscopy (TEM) 6 μL of sample was applied to a glow discharged carbon coated electron microscopy grid (CF150-Cu, Electron Microscopy Sciences) for 5 minutes. Then grids were stained with 2% uranyl acetate for 2 minutes. Samples were visualized using a FEI Tecnai T12 Quick room temperature transmission electron microscope equipped with a Gatan 2,048 x 2,048 CCD camera operated at an acceleration voltage of 120 kV. AD brain tau fibril seeding in tau biosensor cell line HEK293 cell lines stably expressing tau-K18-YFP were engineered by Marc Diamond’s laboratory at the University of Texas Southwestern Medical Center and used without further characterization or authentication. Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies, cat# 11965092) supplemented with 10% (v/v) fetal bovine serum (Life Technologies, cat# A3160401), 1% antibiotic-antimycotic (Life Technologies, cat# 15140122), and 1% Glutamax (Life Technologies, cat# 35050061) at 37 °C, 5% CO 2 in a humidified incubator. AD-tau fibrils were incubated with D-TLKIVWX (X = A, S, D, I, V, R, K, E, P, T) (2.5, 5, 10, 20, 50, 75, 100 μM) at 4 °C overnight and sonicated in a cup horn water bath for 3 min. Then these disassembly products of AD-tau were mixed with 1 volume of Lipofectamine 3000 prepared by diluting 1μL of Lipofectamine 3000 (Life Technologies, cat# 2729899) in 19 μL Opti-MEM (Life Technologies, cat# 31985070). After 20 min, 10 μL of fibrils were added to 90 μL tau biosensor cells. After 24 hours of incubation, the number of seeded aggregates was determined by imaging the entire well of a 96-well plate in triplicate using a Celigo image cytometer (Nexcelom) in the YFP channel. The data analysis was described before. For high-quality images, cells were photographed on a ZEISS Axio Observer D1 fluorescence microscope using the EGFP fluorescence channel. Cell toxicity of AD-tau disassembly species AD-tau fibril disassembly species were produced by incubating AD-tau fibrils (estimated 1 μM) with D-TLKIVWX (5, 10, 20, 50, or 100 μM) at 4 °C overnight. Neuro 2A (N2a) cells were cultured in DMEM supplemented with 10% (v/v) fetal bovine serum, 1% antibiotic-antimycotic, and 1% Glutamax in a 5% CO 2 humidified environment at 37 °C. Cells were plated at a density of roughly 6,000 cells/well on 96-well plates in 90 μL of fresh medium. After 24 h, 10 μL of the above AD-tau fibril disassembly species were added and the cells were incubated for another 24 h at 37 °C. Cytotoxicity was measured utilizing an MTT assay. Western blot Purified brain-extracted AD-tau fibrils were incubated with 500 μM D-TLKIVWI at 37 ℃ for 48 h. The sample was centrifuged at 21,000 x g for 30 min at 4 ℃ (Eppendorf Centrifuge 5424R). Western blot was performed with anti-tau rabbit polyclonal antibody (1:1000, Dako A0024) and anti-rabbit HRP-conjugated secondary antibody (1:5000, Thermo Fisher Scientific). X-ray diffraction (XRD) D-TLKIVWX (X = I, S, R) peptides were dissolved to 10 mM in deionized water and incubated at room temperature quiescently for three days. Peptide fibrils were aligned by pipetting the suspension in a 3 mm gap between two fire-polished glass rods and drying overnight. The aligned fibrils were cooled to 100 K. Diffraction data was collected on a FR-E+ rotating anode x-ray generator (Rigaku) equipped with a R-AXIS HTC imaging plate detector (Rigaku). Cu K-α x-ray beam with 1.5406 Å wavelength was used, and the detector was placed 78 mm from the sample. Diffraction images were visualized using ADXV (The Scripps Research Institute). Atomic force microscopy (AFM) 4 mM D-TLKIVWI in deionized water was shaken at room temperature for three days and diluted into distilled water in a 1:10 ratio. Then, 5 μL of diluted sample was deposited onto freshly cleaved mica and incubated for 10 min. The sample was rinsed with Milli-Q water and dried under a stream of nitrogen gas. AFM images were collected using a Dimension Icon microscope (Bruker) in PeakForce Tapping mode using ScanAsyst-HR probes. Each collected image had a scan size of 3 x 3 μm and 2048 x 2048 pixels and was collected using a scan rate of 0.494 Hz. Nanoscope Analysis software (Version 2.0, Bruker) was used to process the image data by flattening the height topology data to remove tilt and scanner bow. Fibrils were traced and computationally straightened from collected AFM images in Matlab using Trace_y 35 . Cryo-EM samples D-TLKIVWI fibrils were optimized by shaking 4 mM D-TLKIVWI in deionized water at room temperature for three days. 10 mM D-TLKIVWS/R in deionized water formed fibrils when left undisturbed for three days at room temperature. For D-TLKIVWR fibrils, the pH of the peptide solution was adjusted to 7.0. Prior to cryo-EM grid preparation, AD-tau fibrils in a buffer comprised of 20 mM Tris-HCl pH 7.4, 100 mM NaCl were pre-incubated at 37 °C with final concentration of 100 µM D-TLKIVWX (X = I, S, R) from 10 mM stocking solution in water for 24 hours. Control tau fibrils from the same brain donor were treated identically except for the addition of D-TLKIVWX. Cryo-EM data collection and processing To prepare the cryo-EM grids, we applied 2.5 μl of sample solution onto Quantifoil 1.2/1.3 200 mesh electron microscope grids glow-discharged for 2 minutes in a Pelco easiGlow unit before use. Grids were plunge-frozen into liquid nitrogen-cooled liquid ethane inside a Vitrobot Mark IV (FEI) vitrification robot after blotting. Cryo-EM data of D-TLKIVWR and D-TLKIVWS fibrils were collected on a Titan Krios transmission electron microscope (Thermo Fisher Scientific) located at the National Center for Cryo-EM Access and Training, which is equipped with a Bioquantum/K3 direct detection camera (Gatan), operated with 300 kV acceleration voltage and an energy slit width of 20 eV, automated with Leginon software package 36 . Super-resolution movies were collected with a calibrated pixel size of 1.067 Å/pixel (0.5335 Å/pixel in super-resolution movie frames) and a dose per frame of ~1.5 e-/Å 2 . A total of 40 frames with a frame rate of 12 Hz were taken for each movie, resulting in a final dose of ~60 e-/Å 2 per image. D-TLKIVWI fibrils were collected on a Titan Krios located at the HHMI Janelia Research Campus, which is equipped with a cold-FEG source (CFEG), a Selectris X energy filter and a Falcon 4i direct detection camera (TFS), operated with 300 kV acceleration voltage and an energy slit width of 6 eV, and automated with the SerialEM software package 37 . Electron Event Representation (EER) files were collected with a calibrated pixel size of 0.94 Å/pixel and a dose per raw frame of 0.0244 e-/Å 2 , resulting in 55 e-/Å 2 per image. AD-tau/D-TLKIVWX (X = I, S and R) were collected similarly as D-TLKIVWI fibrils, although manually targeted in SerialEM package. The AD-tau control was collected on a Titan Krios/Bioquantum/K3 setup located at the Stanford-SLAC Cryo-EM Center, operated with 300 kV acceleration voltage and an energy slit width of 20 eV, automated with EPU (TFS). (See details in Supplementary Table 1,2). Movies and EER files were motion-corrected in RELION 38 and binned to pixel sizes according to Supplementary Table 1,2. CTF estimation was performed using CTFFIND4 39 . AD-tau/D-TLKIVWX fibrils were manually picked using e2helixboxer.py from EMAN2 40 . D-TLKIVWX fibrils and AD-tau control particle picking was initially done manually using e2helixboxer.py from EMAN2 for about 100 images as a training set for crYOLO 41 . CrYOLO was then trained with default parameters and was used to pick the rest of the images. Particle extraction, two-dimensional classification, three-dimensional classification, and 3D refinement were performed in RELION 42 . Briefly, particles were initially extracted using a larger box size of 640 pixels with two-fold binning. 2D classification was then performed with all particles to eliminate bad particles and group particles into polymorphs if necessary. Particles from each polymorph were selected, extracted with smaller box sizes at native pixel sizes of detectors (binning=1), further “purified” using 2D classification and subjected to 3D classification, which was done initially with one class and then with three classes, using a Gaussian cylinder as the initial model. The best 3D classes were selected, and corresponding particles were finally refined with 3D auto-refine for the reported maps. (See details in Table Supplementary Table 1,2). Part of the Cryo-EM data processing used Expanse GPU at San Diego Supercomputer Center through allocation BIO230174 from the Advanced Cyberinfrastructure Coordination Ecosystem 43 . Atomic model building Our starting atomic model of D-TLKIVWI was an ideal β-strand. It was manually adjusted to fit the electrostatic potential map using Coot 44 and automated refinement was performed using Phenix 45 . To facilitate good rotamer geometry, the initial building and refinement was performed using a map with handedness chosen so that the amino acid residues appeared to be levorotary rather than dextrorotary. In this way, we could take advantage of the rotamer library in Coot which exists for L-amino acids, but not for D-amino acids. In the final step of refinement, the map and coordinates were inverted to the correct hand, consistent with D-amino acids. The starting models for D-TLKIVWS and D-TLKIVWR were adapted from the refined D-TLKIVWI structure. All atomic models were refined in successive rounds using Coot for manual building and Phenix for automated refinement. Model validation statistics of all three D-peptide structures are reported in Supplementary Table 1. Our starting atomic model of the complex between AD-tau PHF and D-TLKIVWI was built by manually orienting coordinates of the tau PHF (PDB ID 7nrv) 46 to fit the electrostatic potential map using Coot 44 and then refined with Phenix. Coordinates of a pair of β-sheets were extracted from the D-TLKIVWI structure described above and manually docked on the surface of the PHF using guidance from the 3.1 Å cryoEM map, as well as the low-pass filtered map (7 Å). We noted a blob of residual density situated at the end of three lysine side chains: K317 and K321 of tau and K3 of the D-peptide. Whatever molecule produced this residual density does not depend on the presence of the D-peptide to bind to tau, since a similar blob was evident in our PHF negative control lacking D-peptide. Indeed, the presence of this residual density was noted in the original structure report of AD-tau PHFs 30 , and even noted in maps from PHFs produced with recombinant tau (PDB ID 7ql4) 47 . The chemical environment of this blob suggests that the blob originates from an anion, but the density is not sufficiently detailed to uniquely identify the chemical species. It is roughly the size of a pair of phosphate ions. We chose to model this residual density with ethylenediaminetetraacetate (EDTA) because it fits the density, caries the expected negative charges to complement the positive charge on K317 and K321, and we know that EDTA was included in the buffer used for PHF purification. The starting models for tau complexed with D-TLKIVWS and D-TLKIVWR were obtained using an analogous procedure. The final refined coordinates for these two complexes do not include the D-peptides because density for the peptides was visible only in the low-pass filtered maps, and not in the high-resolution map (Figure 4 and Extended data Figure 9a). NMR spectroscopy NMR samples were ~0.5 mL of 0.1 mM 15 N, 13 C-labeled tau K18+ protein in 100 mM KCl, 20 mM NaH 2 PO 4 , 1 mM TCEP, 5%/95% D 2 O/H 2 O, pH 7.0 without or with 1 mM D-TLKIVWI, D-TLKIVWS, L-TLKIVWI. All NMR spectra were acquired at 25 °C with Bruker Avance III HD 600 MHz spectrometer equipped with QCI HCNP cryoprobe or Avance Neo 800 MHz spectrometer equipped with TCI HCN cryoprobe. Backbone assignments for both free tau K18+ and D-TLKIVWI bound tau K18+ were carried out using HNCACB, CBCA(CO)NH and C(CO)NH NMR experiments. NMR spectra were acquired with Topspin (Bruker), processed with NMRPipe 48 , and analyzed with NMRFAM-Sparky 49 . For the chemical shift perturbation (CSP) analysis, the overall change in chemical shift Δ was calculated between the free and bound states of tau K18+ protein as 50 : where Δδ H and Δδ N are the differences between the 1 H N and 15 N chemical shifts of the two states being compared. Statistical analysis Graphs are expressed as means + standard deviation (SD) and data were analyzed using SPSS 25 statistical analysis software (SPSS, Chicago, IL, USA). The one-way analysis of variance (ANOVA) was used to analyze difference among multiple groups. Statistical differences for all tests were considered significant at the * p < 0.05, ** p < 0.01, *** p < 0.001 levels, NS, not significant. Abbreviations Aβ, amyloid beta protein; AD, Alzheimer’s disease, AD-tau, tau fibrils extracted from AD brains; All D-peptides TLKIVWX (D-TLKIVWX), in which X can be any enantiomorph of the 20 coded amino acid residues; PHFs, paired helical filaments, the most abundant polymorph of AD-tau fibril. Declarations Reporting summary Further information on research design is available in the Nature Research Reporting Summary linked to this paper. Data availability Cryo-EM maps and atomic models of the fibrils from peptides D-TLKIVWX (X = I, S, R) alone have been deposited in the Electron Microscopy Data Bank (EMDB) and PDB under the accession codes EMD-44181/9B4I (D-TLKIVWI), EMD-44182/9B4J (D-TLKIVWS), and EMD-44183/9B4K (D-TLKIVWR). Cryo-EM maps and atomic models of paired helical filament of Tau from Alzheimer’s patient in complex with the three peptides have been deposited in the same databases under the accession codes EMD-44184/9B4L (Tau-D-TLKIVWI), EMD-44185/9B4M (Tau-D-TLKIVWS), and EMD-44186/9B4N (Tau-D-TLKIVWR). Cryo-EM map and atomic model of the paired helical filament from the same patient have been deposited under the accession codes EMD-44187/9B4O. All data presented in this article are available within the figures and Supplementary Information files. All other data are available from the corresponding authors upon reasonable request. Acknowledgment s We thank the donors and their families for the AD brain tissue; without whom this work would not have been possible. We acknowledge NIH 1R01AG070895 (D.S.E.), NIH RF1AG065407 (D.S.E.), DOE-FC02-02ERG and Alzheimer's Association Research Fellowship AARF-21-848751 (K.H.) for support. The authors thank Dr. Marc Diamond for sharing the YFP-labeled tau K18 expressing HEK293 cells. We thank Virginia Lee for generously gifting the GT38 antibody. Some of this work was performed at the National Center for CryoEM Access and Training (NCCAT) and the Simons Electron Microscopy Center located at the New York Structural Biology Center, supported by the NIH Common Fund Transformative High Resolution Cryo-Electron Microscopy program (U24 GM129539), and by grants from the Simons Foundation (SF349247) and NY State Assembly. We also thank the staff at the HHMI Janelia CryoEM Facility for help and support. Some of this work was performed at the Stanford-SLAC Cryo-EM Center (S2C2) supported by the NIH Common Fund Transformative High Resolution Cryo-Electron Microscopy program (U24 GM129541). This work used Expanse GPU at San Diego Supercomputer Center through allocation BIO230174 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296. We acknowledge support of NMR equipment by grants from NIH (S10OD016336 and S10OD025073) and from DOE (DE-FC02-02ER63421). Y.Y. was supported in part by R35GM131901 to J.F. Author contributions The project was conceived and designed by K.H. and D.S.E. Cell seeding experiment was performed by J.L.D. and K.H.. Dot blot, TEM and western blot were performed by K.H. and J.P.. J.Z. and J.L.D. provided the recombinant tau K18+. J.P. extracted tau fibrils from AD patient brains with the help of X.C.. X-ray diffraction was done by K.H. and Y.X.J.. AFM was carried out by L.L.. Cryo-EM grids were prepared by K.H., Y.X.J. and P.G.. Cryo-EM data were collected by P.G. with assistance from S.Y., Z.Y., Y.X.J. and D.R.B.. K.H. and P.G. processed cryo-EM data with assistance from Y.X.J., D.R.B., X.C. and J.L.. M.R.S. and K.H. built atomic models. Isotopically labeled tau K18+ was purified by K.H. NMR experiments were conducted and analyzed by Y.Y. and J.F.. The manuscript was prepared by K.H., M.R.S., P.G., and D.S.E. with contributions from all the other authors. Competing interests D.S.E. is SAB chair and equity holder of ADRx, Inc. All other authors declare no conflicts. Part of the work was disclosed in our provisional patent application (Serial No. 63/510,194). References Arriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T. & Hyman, B. T. Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology 42 , 631–631 (1992). Kollmer, M. et al. Cryo-EM structure and polymorphism of Aβ amyloid fibrils purified from Alzheimer’s brain tissue. Nat. Commun. 10 , 4760 (2019). Scheres, S. H. W., Ryskeldi-Falcon, B. & Goedert, M. Molecular pathology of neurodegenerative diseases by cryo-EM of amyloids. Nature 621 , 701–710 (2023). Low, K. J. Y., Venkatraman, A., Mehta, J. S. & Pervushin, K. Molecular mechanisms of amyloid disaggregation. J. Adv. Res. 36 , 113–132 (2022). Iadanza, M. G., Jackson, M. P., Hewitt, E. W., Ranson, N. A. & Radford, S. E. A new era for understanding amyloid structures and disease. Nat. Rev. Mol. Cell Biol. 19 , 755–773 (2018). Nachman, E. et al. Disassembly of Tau fibrils by the human Hsp70 disaggregation machinery generates small seeding-competent species. J. Biol. Chem. 295 , 9676–9690 (2020). Al-Hilaly, Y. K. et al. Cysteine-Independent Inhibition of Alzheimer’s Disease-like Paired Helical Filament Assembly by Leuco-Methylthioninium (LMT). J. Mol. Biol. 430 , 4119–4131 (2018). Low, K. J. Y., Venkatraman, A., Mehta, J. S. & Pervushin, K. Molecular mechanisms of amyloid disaggregation. J. Adv. Res. 36 , 113–132 (2022). Seidler, P. M. et al. Structure-based discovery of small molecules that disaggregate Alzheimer’s disease tissue derived tau fibrils in vitro. Nat. Commun. 13 , 5451 (2022). Wischik, C. M., Harrington, C. R. & Storey, J. M. Tau-aggregation inhibitor therapy for Alzheimer’s disease. Biochem. Pharmacol. 88 , 529–539 (2014). Sievers, S. A. et al. Structure-based design of non-natural amino-acid inhibitors of amyloid fibril formation. Nature 475 , 96–100 (2011). Chong, P., Sia, C., Tripet, B., James, O. & Klein, M. Comparative immunological properties of enantiomeric peptides. Lett. Pept. Sci. 3 , 99–106 (1996). Peters, O., Cosma, N. C., Kutzsche, J. & Willbold, D. A randomized, placebo‐controlled, double‐blind, Phase 1b study to evaluate the safety, tolerability and pharmacodynamics of PRI‐002 in early AD. Alzheimers Dement. 18 , e069253 (2022). Hou, K. et al. D-peptide-magnetic nanoparticles fragment tau fibrils and rescue behavioral deficits in a mouse model of Alzheimer’s disease. Sci. Adv. (2024), in press. Weismiller, H. A., Holub, T. J., Krzesinski, B. J. & Margittai, M. A thiol-based intramolecular redox switch in four-repeat tau controls fibril assembly and disassembly. J. Biol. Chem. 297 , (2021). Prifti, E. et al. The two cysteines of tau protein are functionally distinct and contribute differentially to its pathogenicity in vivo. J. Neurosci. 41 , 797–810 (2021). Kumat, T., Samuel, D., Jayaraman, G., Srimathi, T. & Yu, C. The role of proline in the prevention of aggregation during protein folding in vitro. IUBMB Life 46 , 509–517 (1998). Gibbons, G. S. et al. Detection of Alzheimer’s disease (AD) specific tau pathology with conformation-selective anti-tau monoclonal antibody in co-morbid frontotemporal lobar degeneration-tau (FTLD-tau). Acta Neuropathol. Commun. 7 , 1–12 (2019). Willbold, D., Strodel, B., Schröder, G. F., Hoyer, W. & Heise, H. Amyloid-type Protein Aggregation and Prion-like Properties of Amyloids. Chem. Rev. 121 , 8285–8307 (2021). Holmes, B. B. et al. Proteopathic tau seeding predicts tauopathy in vivo. Proc. Natl. Acad. Sci. 111 , E4376–E4385 (2014). Serpell, L. C. Alzheimer’s amyloid fibrils: structure and assembly. Biochim. Biophys. Acta BBA-Mol. Basis Dis. 1502 , 16–30 (2000). Hughes, E., Burke, R. M. & Doig, A. J. Inhibition of toxicity in the β-amyloid peptide fragment β-(25–35) using N-methylated derivatives: a general strategy to prevent amyloid formation. J. Biol. Chem. 275 , 25109–25115 (2000). Wang, M. et al. Left or Right: How Does Amino Acid Chirality Affect the Handedness of Nanostructures Self-Assembled from Short Amphiphilic Peptides? J. Am. Chem. Soc. 139 , 4185–4194 (2017). Ke, P. C. et al. Half a century of amyloids: past, present and future. Chem. Soc. Rev. 49 , 5473–5509 (2020). Adamcik, J. & Mezzenga, R. Study of amyloid fibrils via atomic force microscopy. Curr. Opin. Colloid Interface Sci. 17 , 369–376 (2012). Lutter, L. et al. Structural Identification of Individual Helical Amyloid Filaments by Integration of Cryo-Electron Microscopy-Derived Maps in Comparative Morphometric Atomic Force Microscopy Image Analysis. J. Mol. Biol. 434 , 167466 (2022). Close, W. et al. Physical basis of amyloid fibril polymorphism. Nat. Commun. 9 , 699 (2018). Nelson, R. et al. Structure of the cross-β spine of amyloid-like fibrils. Nature 435 , 773–778 (2005). Sawaya, M. R. et al. Atomic structures of amyloid cross-β spines reveal varied steric zippers. Nature 447 , 453–457 (2007). Fitzpatrick, A. W. et al. Cryo-EM structures of tau filaments from Alzheimer’s disease. Nature 547 , 185–190 (2017). Barré, P. & Eliezer, D. Structural transitions in tau k18 on micelle binding suggest a hierarchy in the efficacy of individual microtubule‐binding repeats in filament nucleation. Protein Sci. 22 , 1037–1048 (2013). Cribbs, D. H., Pike, C. J., Weinstein, S. L., Velazquez, P. & Cotman, C. W. All-D-enantiomers of β-amyloid exhibit similar biological properties to all-L-β-amyloids. J. Biol. Chem. 272 , 7431–7436 (1997). Shi, Y. et al. Structure-based classification of tauopathies. Nature 598 , 359–363 (2021). Lu, J. et al. CryoEM structure of the low-complexity domain of hnRNPA2 and its conversion to pathogenic amyloid. Nat. Commun. 11 , 4090 (2020). Xue, W.-F. Trace_y: Software algorithms for structural analysis of individual helical filaments by three-dimensional contact point reconstruction atomic force microscopy. bioRxiv 2023–07 (2023). Suloway, C. et al. Automated molecular microscopy: the new Leginon system. J. Struct. Biol. 151 , 41–60 (2005). Mastronarde, D. N. Automated electron microscope tomography using robust prediction of specimen movements. J. Struct. Biol. 152 , 36–51 (2005). Scheres, S. H. RELION: implementation of a Bayesian approach to cryo-EM structure determination. J. Struct. Biol. 180 , 519–530 (2012). Rohou, A. & Grigorieff, N. CTFFIND4: Fast and accurate defocus estimation from electron micrographs. J. Struct. Biol. 192 , 216–221 (2015). Tang, G. et al. EMAN2: an extensible image processing suite for electron microscopy. J. Struct. Biol. 157 , 38–46 (2007). Wagner, T. et al. Two particle-picking procedures for filamentous proteins: SPHIRE-crYOLO filament mode and SPHIRE-STRIPER. Acta Crystallogr. Sect. Struct. Biol. 76 , 613–620 (2020). He, S. & Scheres, S. H. Helical reconstruction in RELION. J. Struct. Biol. 198 , 163–176 (2017). Boerner, T. J., Deems, S., Furlani, T. R., Knuth, S. L. & Towns, J. Access: Advancing innovation: Nsf’s advanced cyberinfrastructure coordination ecosystem: Services & support. in Practice and Experience in Advanced Research Computing 173–176 (2023). Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60 , 2126–2132 (2004). Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr. Sect. Struct. Biol. 75 , 861–877 (2019). Shi, Y. et al. Cryo-EM structures of tau filaments from Alzheimer’s disease with PET ligand APN-1607. Acta Neuropathol. (Berl.) 141 , 697–708 (2021). Lövestam, S. et al. Assembly of recombinant tau into filaments identical to those of Alzheimer’s disease and chronic traumatic encephalopathy. Elife 11 , e76494 (2022). Delaglio, F. et al. NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6 , 277–293 (1995). Lee, W., Tonelli, M. & Markley, J. L. NMRFAM-SPARKY: enhanced software for biomolecular NMR spectroscopy. Bioinformatics 31 , 1325–1327 (2015). Williamson, M. P. Using chemical shift perturbation to characterise ligand binding. Prog. Nucl. Magn. Reson. Spectrosc. 73 , 1–16 (2013). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTable1CryoEMstatisticsofDTLKIVWX.doc Supplementary Table 1 SupplementaryTable2CryoEMstatisticsofADtauDTLKIVWX.doc Supplementary Table 2 SIGuide.pdf SIGuide TheStrainReliefmechanismforthedisassemblyofAlzheimersDiseasefibrilsofproteintaubyDpeptideDTLKIVWI.mp4 Strain-relief of D-TLKIVWX fibrils drives disassembly of AD-tau fibrils. ExtendedDataFigs.docx Cite Share Download PDF Status: Published Journal Publication published 09 Jul, 2025 Read the published version in Nature → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4152095","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":283628091,"identity":"c3914a0a-9a8a-4b97-b50d-fea5c908b1d2","order_by":0,"name":"David Eisenberg","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyUlEQVRIiWNgGAWjYDCCA0DEU8GQwAbiPICKEKHlDFRLArFaGHjbIIqJ08J3/IzhgbfztuXxMfAYfkioYZDju5GAX4vkmRyDg3O33S5mY+Axlkg4xmAsSUiLwYG0hMO8224ntjHwmIF8lLiBoJbzz4Ba5sC0/GOoJ6zlRvKBw7wNUC1AMsGAoF9uPD5wcM4xoF+Y2YolEvskDGeeeYBfC9/5xOYPb2pu58m3N2/88OGbjTzfcQK2IAAzmJQgVvkoGAWjYBSMAnwAABh6StJNk/PBAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-2432-5419","institution":"UCLA","correspondingAuthor":true,"prefix":"","firstName":"David","middleName":"","lastName":"Eisenberg","suffix":""},{"id":283628092,"identity":"1b08f74e-c6bd-4678-8cf3-0e3bac328a49","order_by":1,"name":"Ke Hou","email":"","orcid":"https://orcid.org/0000-0003-3869-2076","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Hou","suffix":""},{"id":283628093,"identity":"0a4a09f6-cf83-48c7-8c38-d6259a423236","order_by":2,"name":"Peng Ge","email":"","orcid":"https://orcid.org/0000-0002-2661-8226","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Peng","middleName":"","lastName":"Ge","suffix":""},{"id":283628094,"identity":"ff145fc6-d1a7-47d2-8c70-f494aeae7d02","order_by":3,"name":"Michael Sawaya","email":"","orcid":"https://orcid.org/0000-0003-0874-9043","institution":"University of California Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Sawaya","suffix":""},{"id":283628095,"identity":"c2e3fd4f-ac80-41f6-89e0-a7a6f0c78f30","order_by":4,"name":"Joshua Dolinsky","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"","lastName":"Dolinsky","suffix":""},{"id":283628096,"identity":"33d9fa5d-f86f-4721-87ad-cfaf4a994902","order_by":5,"name":"Yuan Yang","email":"","orcid":"https://orcid.org/0000-0002-9538-6369","institution":"University of California Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Yuan","middleName":"","lastName":"Yang","suffix":""},{"id":283628097,"identity":"a8f1367b-9838-43eb-819b-fb44447c6546","order_by":6,"name":"Yi Xiao Jiang","email":"","orcid":"https://orcid.org/0000-0003-1433-4952","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"Xiao","lastName":"Jiang","suffix":""},{"id":283628098,"identity":"9c2139bf-954c-426d-91bd-6a709691be44","order_by":7,"name":"Liisa Lutter","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Liisa","middleName":"","lastName":"Lutter","suffix":""},{"id":283628099,"identity":"183bf522-bab5-47b8-8790-43e6b57402c5","order_by":8,"name":"David Boyer","email":"","orcid":"https://orcid.org/0000-0002-4487-0230","institution":"University of California Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Boyer","suffix":""},{"id":283628100,"identity":"5d36e9d0-a86e-4091-904e-0dee5f648a76","order_by":9,"name":"Xinyi Cheng","email":"","orcid":"https://orcid.org/0000-0003-1654-4363","institution":"UCLA","correspondingAuthor":false,"prefix":"","firstName":"Xinyi","middleName":"","lastName":"Cheng","suffix":""},{"id":283628101,"identity":"e2136105-ef85-42d7-b480-af1fdfbaaaa6","order_by":10,"name":"Justin Pi","email":"","orcid":"https://orcid.org/0009-0000-9956-6351","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Justin","middleName":"","lastName":"Pi","suffix":""},{"id":283628102,"identity":"d92cfd79-2c99-445a-a9da-2ce7e9607541","order_by":11,"name":"Jeffrey Zhang","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Jeffrey","middleName":"","lastName":"Zhang","suffix":""},{"id":283628103,"identity":"d70490df-fbd9-40e1-b6ae-1d0492c578e3","order_by":12,"name":"Jiahui Lu","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Jiahui","middleName":"","lastName":"Lu","suffix":""},{"id":283628104,"identity":"353da53d-c446-42bb-a9d1-6a9a95e62971","order_by":13,"name":"Shixin Yang","email":"","orcid":"https://orcid.org/0000-0002-9794-7537","institution":"Janelia Research Campus, Howard Hughes Medical Institute","correspondingAuthor":false,"prefix":"","firstName":"Shixin","middleName":"","lastName":"Yang","suffix":""},{"id":283628105,"identity":"7f035a18-318c-4a98-abc4-66a8b0f28f58","order_by":14,"name":"Zhiheng Yu","email":"","orcid":"https://orcid.org/0000-0002-7114-2570","institution":"Howard Hughes Medical Institute","correspondingAuthor":false,"prefix":"","firstName":"Zhiheng","middleName":"","lastName":"Yu","suffix":""},{"id":283628106,"identity":"8fa83cb9-2b0c-46bc-b5c0-cfce3cd20c11","order_by":15,"name":"Juli Feigon","email":"","orcid":"https://orcid.org/0000-0003-3376-435X","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Juli","middleName":"","lastName":"Feigon","suffix":""}],"badges":[],"createdAt":"2024-03-22 22:45:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4152095/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4152095/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41586-025-09244-z","type":"published","date":"2025-07-09T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":56118407,"identity":"0970ddf9-f7c2-4051-b934-38569623ce9f","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":982352,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eD-TLKIVWX peptides disassemble AD-tau fibrils to non-seeding products with varying efficiency. a, \u003c/strong\u003eDot blot staining of AD-tau fibrils before and after incubation with 500 μM D-TLKIVW, D-TLKIVWX (X = C, A, S, D, I, V, R, K, E, P, T) at 37 ℃ for 48 h, probed by the GT38 antibody. Each sample was analyzed in triplicate. \u003cstrong\u003eb\u003c/strong\u003e, Quantification of the level of AD-tau fibrils in panel a by ImageJ. Statistical significance was analyzed by one-way ANOVA (*\u003cem\u003ep \u003c/em\u003e\u0026lt; 0.05; the \u003cem\u003ep\u003c/em\u003e vale is 0.026, 0.013, 0.002, 0.000, 0.001, 0.000, 0.000, 0.000 between AD-tau group and D-TKLIVWX (X = K, R, S, T, C, A, V, I) group, respectively). \u003cstrong\u003ec\u003c/strong\u003e, Representative TEM images of AD-tau fibrils after incubation with 500 μM D-TLKIVWX (X = I, S, R) at 37 ℃ for 48 h. The AD-tau fibrils and newly formed fibrils are labeled by red and blue arrows, respectively. Scale bars, 100 nm. \u003cstrong\u003ed\u003c/strong\u003e, Representative fluorescent images of HEK293 cells expressing YFP-labeled tau K18 transfected with AD-tau fibril seeds after overnight disassembly with various concentrations of D-TLKIVWX (X = R, S, I). Fluorescent puncta are marked by white arrows. Scale bars, 100 μm. \u003cstrong\u003ee\u003c/strong\u003e, Cell viability of N2a cells treated with AD-tau fibrils in various concentrations of D-TLKIVWI measured by MTT assay. Results shown as Mean + SD of triplicate wells. Statistical significance was analyzed by one-way ANOVA (*\u003cem\u003ep \u003c/em\u003e\u0026lt; 0.05; the \u003cem\u003ep\u003c/em\u003e vale is 0.322, 0.021, 0.008, 0.009, 0.000 between 0 µM group and 5, 10, 20, 50 or 100 µM group, respectively).\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/ebf4e7631563626a125502a5.png"},{"id":56118406,"identity":"07e91b6e-67cc-4030-b52b-239da20fc22d","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1467476,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eD-TLKIVWX disassembly of AD-tau fibrils involves formation of D-TLKIVWX amyloid-like fibrils.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eNegative stain transmission electron micrographs of AD-tau fibrils incubated with 500 μM D-TLKIVWI at 0, 1, 3, 6, 24 and 48 h timepoints. AD-tau fibrils, newly formed D-TLKIVWI fibrils, and amorphous products are labeled by red, blue and magenta arrows, respectively. \u003cstrong\u003eb\u003c/strong\u003e, Negative stain transmission electron micrograph of D-TKLIVWI amyloid-like fibrils formed by 10 mM D-TLKIVWI grown under quiescent conditions in deionized water at room temperature for three days. \u003cstrong\u003ec\u003c/strong\u003e, X-ray diffraction pattern of D-TLKIVWI fibrils showing spacings characteristic of amyloid-like fibrils. \u003cstrong\u003ed\u003c/strong\u003e, Negative stain transmission electron micrograph of 500 μM D-TLK(N-Me-I)VWI incubated in 20 mM Tris-HCl, pH 7.4, 100 mM NaCl at 37 ℃ for 48 h. \u003cstrong\u003ee\u003c/strong\u003e, Representative TEM image of AD-tau fibrils after incubation with 500 μM D-TLK(N-Me-I)VWI at 37 ℃ for 48 h\u003cstrong\u003e.\u003c/strong\u003e AD-tau fibrils are labeled by red arrows. \u003cstrong\u003ef\u003c/strong\u003e, Dot blot staining of AD-tau fibrils before and after incubation with 500 μM D-TLKIVWX or D-TLK(N-Me-I)VWX (X = I, S and R) at 37 ℃ for 48 h, probed by the GT38 antibody. \u003cstrong\u003eg\u003c/strong\u003e, Quantification of the dot blot staining in panel f. Results shown as Mean + SD of triplicate dots. Statistical significance was analyzed by one-way ANOVA (***\u003cem\u003ep \u003c/em\u003e\u0026lt; 0.001; NS, not significant; the \u003cem\u003ep\u003c/em\u003e vale is 0.000, 0.000, 0.000, 0.869, 0.999, 0.263 between AD tau group and D-TLKIVWX (X= I, S, or R) or D-TLK(N-Me-I)VWX (X = I, S or R) group, respectively).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/d859217c2b744b3abe60b2b3.png"},{"id":56118412,"identity":"344e625e-87ff-4350-9b1e-31b7a640c980","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":809464,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eD-TLKIVWX (X = I, S, R) peptides form right-handed helical fibrils with conserved steric zipper motifs. a\u003c/strong\u003e, Representative atomic force microscopy (AFM) image of D-TLKIVWI fibrils.\u003cstrong\u003e b-d, \u003c/strong\u003eViews down the fibril axes (upper row) and perpendicular to the fibril axes (lower row)\u003cstrong\u003e \u003c/strong\u003eof the amyloid-like fibrils formed by (\u003cstrong\u003eb\u003c/strong\u003e) D-TLKIVWI, (\u003cstrong\u003ec\u003c/strong\u003e) D-TLKIVWS, (\u003cstrong\u003ed\u003c/strong\u003e) D-TLKIVWR, as determined by cryo-EM. The conserved steric zipper motifs are represented by grey dotted ellipses.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/b3d73d692607f500e01ecb74.png"},{"id":56118411,"identity":"02487704-8a43-4c0c-80e2-b7f613dd25f6","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":679182,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCryo-EM structures of D-TLKIVWX (X = I, S and R) in the act of disassembling tau PHF. a\u003c/strong\u003e-\u003cstrong\u003ec\u003c/strong\u003e, Cryo-EM density, and atomic models of a cross section of PHF fibrils complexed with (\u003cstrong\u003ea\u003c/strong\u003e) D-TLKIVWI, (\u003cstrong\u003eb\u003c/strong\u003e), D-TLKIVWS, and (\u003cstrong\u003ec\u003c/strong\u003e), D-TLKIVWR. Details of the PHF structures are evident in sharpened, high-resolution density maps of the PHFs (dark salmon). The D-peptides are visible in unsharpened, low-pass filtered density (7 Å) colored blue (D-TLKIVWI), marine (D-TLKIVWS), and green (D-TLKIVWR). The corresponding low-pass filtered density of the PHF is shown in light salmon. A small blob of residual density sandwiched between the PHF and each of the D-peptides was modeled with EDTA. \u003cstrong\u003ed\u003c/strong\u003e, \u003csup\u003e1\u003c/sup\u003eH-\u003csup\u003e15\u003c/sup\u003eN HSQC spectral overlay of free \u003csup\u003e15\u003c/sup\u003eN,\u003csup\u003e13\u003c/sup\u003eC-labeled tau K18+ (salmon), and \u003csup\u003e15\u003c/sup\u003eN,\u003csup\u003e13\u003c/sup\u003eC-labeled tau K18+ with 10-fold molar excess of D-TLKIVWI after incubating for 2.5 months at 4 °C (blue). Representative residues with significant chemical shift changes are labeled and resonance shifts indicated with arrows. These shifted residues (Val309-Thr319) are highlighted in yellow in panel a.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/03e378086b08534e137947c8.png"},{"id":56118798,"identity":"6ea8a811-5693-4c51-9b88-1d76f159e434","added_by":"auto","created_at":"2024-05-08 19:14:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":571661,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProposed strain-relief mechanism of disassembly of tau PHF by D-TLKIVWX peptide. a\u003c/strong\u003e, Diagram of left-handed tau PHF. \u003cstrong\u003eb\u003c/strong\u003e, D-TLKIVWX peptides assemble along the axis of tau PHF forming β-sheet-β-sheet interactions stabilized by steric zippers. This produces a metastable D-TLKIVWX protofilament with the left-twisted handed twist of the tau PHF. The strained left-twisted D-TLKIVWX protofilament is at a higher potential energy relative to the right-twist preferred by D-TLKIVWX peptides. \u003cstrong\u003ec,\u003c/strong\u003e Torsional strain is relieved as the D-TLKIVWX protofilament relaxes to a right-handed twist. The tau PHF sheet, in contrast, develops new strain as it struggles to maintain its interface with the now right-twisted D-TLKIVWX protofilament. Backbone hydrogen-bonds break in the tau PHF at the junction between left-and right twisted segments (red dotted lines).\u003cstrong\u003e d\u003c/strong\u003e, Solvent invasion of the tau PHF core facilitates further disassembly of the PHF. The disassembly products of the reaction are complexes of tau and D-TLKIVWX.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/3913e7678b3e74d7a0fe1614.png"},{"id":86402694,"identity":"c631b43e-c7d3-4390-aaef-5dc8613fb28b","added_by":"auto","created_at":"2025-07-10 09:10:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5837211,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/eccab83f-a4a2-4069-bedb-2aad6b2a801c.pdf"},{"id":56118405,"identity":"61c16d1d-8c94-4eb4-848a-e7db2d3570fa","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":71168,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table 1\u003c/p\u003e","description":"","filename":"SupplementaryTable1CryoEMstatisticsofDTLKIVWX.doc","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/cfe849a004fff4f145347019.doc"},{"id":56118404,"identity":"894f67af-fb1f-4c7b-a24d-fa4287362a7c","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"doc","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":80896,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table 2\u003c/p\u003e","description":"","filename":"SupplementaryTable2CryoEMstatisticsofADtauDTLKIVWX.doc","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/3c370d78bb13a2e42322e1ca.doc"},{"id":56118408,"identity":"5166194e-9047-4a44-a356-fc6894c052aa","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":106272,"visible":true,"origin":"","legend":"SIGuide","description":"","filename":"SIGuide.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/8a5ca1cadcc10dbca60ba0b5.pdf"},{"id":56118414,"identity":"0b335812-67d7-422e-9e98-fa7a838bd3ad","added_by":"auto","created_at":"2024-05-08 19:06:27","extension":"mp4","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":196675690,"visible":true,"origin":"","legend":"Strain-relief of D-TLKIVWX fibrils drives disassembly of AD-tau fibrils.","description":"","filename":"TheStrainReliefmechanismforthedisassemblyofAlzheimersDiseasefibrilsofproteintaubyDpeptideDTLKIVWI.mp4","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/2518af24baf6021c6271c979.mp4"},{"id":56118410,"identity":"09d59c5c-85af-4874-80ba-0b14903daaf1","added_by":"auto","created_at":"2024-05-08 19:06:22","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":9073413,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataFigs.docx","url":"https://assets-eu.researchsquare.com/files/rs-4152095/v1/28f4dd8b6da9d30a3b95a109.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"How short peptides can disassemble ultra-stable tau fibrils extracted from Alzheimer’s disease brain by a strain-relief mechanism","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTau pathology caused by the abnormal aggregation of tau is more strongly correlated with cognitive symptoms and severity in Alzheimer\u0026apos;s disease (AD) than A\u0026beta; plaques\u003csup\u003e1\u0026ndash;3\u003c/sup\u003e. Thus, disrupting tau aggregates emerges as a promising therapeutic strategy for AD\u003csup\u003e4,5\u003c/sup\u003e. To date, several types of disaggregators of tau fibrils have been reported and their disaggregation mechanisms are being sought. For example, the human Hsp70/DnaJ system can disaggregate tau fibrils \u003cem\u003ein vitro\u003c/em\u003e in an ATP-dependent manner, but it generates small, seeding-competent species that accelerate the progression of disease\u003csup\u003e6\u003c/sup\u003e. Small molecules such as methylene blue\u003csup\u003e7\u003c/sup\u003e and EGCG, by binding to tau fibrils, apparently disrupt intermolecular interactions\u003csup\u003e8\u003c/sup\u003e. Our recent structural studies revealed EGCG stacks in the clefts formed at the junction of the two protofilaments in AD tau fibrils\u003csup\u003e9\u003c/sup\u003e. However, small molecules including methylene blue/LMTX and EGCG have not been proven to be effective drugs, perhaps because of limited bioavailability, promiscuous protein binding and low blood-brain barrier permeability\u003csup\u003e10\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eA peptide-based disaggregator can offer advantages over small molecules including higher specificity and higher binding affinity\u003csup\u003e11\u003c/sup\u003e. Especially, compared to L-peptides, D-enantiomeric peptides are known to be less immunogenic, less protease-sensitive\u003cem\u003e\u0026nbsp;in vitro\u003c/em\u003e and more resistant to degradation \u003cem\u003ein vivo\u003c/em\u003e\u003csup\u003e12\u003c/sup\u003e\u003cem\u003e.\u003c/em\u003e A recent phase I clinical trial of a D-peptide that disassembles A\u0026beta; oligomers has proved to be safe and well tolerated\u003csup\u003e13\u003c/sup\u003e. Previously, we reported that the D-peptide D-TLKIVWC can disassemble\u0026nbsp;tau fibrils extracted from AD brains (AD-tau),\u0026nbsp;neutralizing their seeding ability and rescuing behavioral deficits in a mouse model of Alzheimer\u0026rsquo;s disease\u003csup\u003e14\u003c/sup\u003e. However, the underlying disassembly mechanism remains unknown, preventing further development of this type of drug candidate for AD.\u003c/p\u003e\n\u003cp\u003eHere, we designed a series of peptides of sequence D-TLKIVWX varying only at the seventh residue, X. These D-peptides showed variable efficacy in disassembling AD-tau fibrils \u003cem\u003ein vitro\u003c/em\u003e, with X = Ile as the best performer. From electron microscopy, we discovered that D-TLKIVWX peptides form amyloid-like fibrils themselves, and from atomic force microscopy we learned that these fibrils have a right-handed helical twist, in contrast to the left-handed helical twist of AD-tau. To learn the molecular interactions between the fibrils that participate in disassembly, we determined the cryo-EM structures of D-TLKIVWX protofilaments bound to tau fibrils of opposing twist. Combining our structural data, we propose a strain-relief mechanism for AD-tau fibril disassembly.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD-TLKIVWX peptides disassemble AD-tau fibrils to non-seeding species\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn previous work, we found that D-TLKIVWC can disassemble tau fibrils, whereas its six-residue analog D-TLKIVW cannot\u003csup\u003e14\u003c/sup\u003e. Here we wondered whether the disassembling ability arises from the capacity of the cysteine to form a disulfide bond with tau residue Cys-322 located in the structured core of tau fibrils. The formation of a disulfide bond could alter the conformation of tau in fibrils, resulting in fibril breakdown\u003csup\u003e15\u003c/sup\u003e\u003csup\u003e,\u003c/sup\u003e\u003csup\u003e16\u003c/sup\u003e. To evaluate this hypothesis, we measured disassembly activity as a function of disulfide bond forming potential which we adjusted by including either glutathione (GSH) or glutathione disulfide (GSSG). As shown in Extended Data Fig. 1a, D-TLKIVWC exhibited equivalent capacity to disassemble recombinant tau K18+ (residues Gln244-Glu380 of 4R tau) fibrils across all tested conditions, indicating the formation of a disulfide bond is irrelevant to the disassembly of tau fibrils. \u003c/p\u003e\n\u003cp\u003eTo determine if the cysteine residue of D-TLKIVWC is essential in disassembling tau fibrils, we substituted the D-cysteine with various other D-amino acid residues, including anionic residues (aspartate (D) and glutamate (E)), cationic residues (arginine (R) and lysine (K)), polar residues (serine (S) and threonine (T)), hydrophobic residues (alanine (A), isoleucine (I), and valine (V)), as well as a \u0026beta;-sheet interrupter (proline (P))\u003csup\u003e17\u003c/sup\u003e. Their performance in disassembling AD-tau fibrils after 48 hours of incubation was evaluated by dot blot and electron microscopy (TEM). The dot blot experiment was conducted with the conformational antibody GT38\u003csup\u003e18\u003c/sup\u003e, which specifically recognizes AD-tau fibrils and does not probe monomeric tau or D-peptide controls (Extended Data Fig. 1b). The D-peptide variants (D-TLKIVWX, X = A, S, D, I, V, R, K, E, T) showed varying efficacy in disassembling AD-tau fibrils depending on the type of residue in the seventh position (Fig. 1a-b). Specifically, hydrophobic residues (X = I, V and A) most significantly reduced the level of AD-tau fibrils, with efficacy decreasing in the following order: hydrophobic \u0026gt; polar (X = S and T) \u0026gt; cationic (X = R and K) \u0026gt; anionic (X = D and E). Lastly, the \u0026beta;-sheet interrupter (X = P) and the deletion of the seventh residue both showed no reduction in the level of AD-tau fibrils. This trend was consistently supported by TEM characterization, where the number of AD-tau fibrils (labeled by red arrows in Fig. 1c and Extended Data Fig. 1c) was quantified after 48 hours of disassembly (Extended Data Fig. 1d). Notably, we observed new fibril species (blue arrows in Fig. 1c and Extended Data Fig. 1c). Furthermore, we observed specificity of D-TLKIVWX in disassembling AD-tau fibrils as it cannot disrupt other amyloid fibrils such as \u0026alpha;-syn fibrils or wild type hnRNPA2 fibrils (Extended Data Fig. 2). In summary, most tested residue types X in D-TLKIVWX demonstrate specific disassembly action against AD-tau fibrils, with hydrophobic residues, especially Ile, being the best.\u003c/p\u003e\n\u003cp\u003eIn addition to uncovering the mechanism of disassembly, it is important to ascertain whether disassembly produces pathologic products that seed the growth of additional fibrils\u003csup\u003e19\u003c/sup\u003e. Therefore, we systematically investigated the seeding capacity of the products of AD-tau fibrils after overnight incubation with variants of D-TLKIVWX in a HEK293T cell line stably expressing yellow fluorescent protein (YFP)-fused tau\u003csup\u003e20\u003c/sup\u003e. As illustrated in Fig. 1d, AD-tau fibrils alone can seed the aggregation of endogenous fluorescent tau, leading to the formation of bright puncta (indicated as white arrows) observable under fluorescence microscopy. In contrast, the overnight disassembly products of AD-tau fibrils treated with D-TLKIVWX gradually lost their seeding ability in a dose-dependent manner (Fig. 1d and Extended Data Fig. 3a). Further, automated image analysis of visible puncta revealed a dependence of the seeding capacity on the seventh amino acid residue within D-TLKIVWX (Extended Data Fig. 3b-f). The observed dependence is consistent with the trends observed in our\u003cem\u003e in vitro\u003c/em\u003e disassembly assays (Fig. 1b and Extended Data Fig. 1c,d). Additionally, because of disassembly activity, D-TLKIVWX (X = I and S) shows a dose-dependent effect in reducing AD-tau toxicity in mouse Neuro 2A (N2a) cells (Fig. 1e and Extended Data Fig. 3g). Our results demonstrate that the products of AD tau fibrils disassembled by D-TLKIVWX are not seeding-competent and are non-toxic.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAmyloid-like fibril formation of D-TLKIVWX is essential for tau disassembly\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo better understand the disassembly process of tau fibrils, we conducted a time-course dot blot and TEM experiment using D-TLKIVWI as a representative of the more efficient fibril disassemblers. Dot blot data (Extended Data Fig. 4a) confirmed that D-TLKIVWI gradually reduced the level of AD-tau fibrils as a function of time. The time-course TEM images in Fig. 2a and Extended Data Fig. 4b show AD-tau fibrils (red arrows) appearing to be covered by unknown species after one hour of incubation with D-TLKIVWI, and additional fibrillar structures (blue arrows) become increasingly evident at three- and six-hour timepoints. At 24 h, amorphous products (magenta arrows) appeared concomitant with the disappearance of AD-tau fibrils and reduction of the unidentified fibril species. The amorphous products appeared more dispersed in micrographs acquired at 48 h, and a western blot showed that the disassembly products of AD-tau fibrils consisted primarily of insoluble species; denatured pelleted material migrated as dimers and other multimers, not as monomeric tau (Extended Data Fig. 4c). Notably, in the TEM images of D-TLKIVWX (X = I, S, R, D, E, K, T, C, A and V)-treated AD-tau samples (Fig. 1c and Extended Data Fig. 1c), we also observed the emergence of new fibrils (labeled by blue arrows) accompanying with the disappearance of AD-tau fibrils. We identified these new species as D-TLKIVWX fibrils since D-TLKIVWX (X = I, S, R, D, E, K, T, C, A and V) exhibited aggregation activity in the same buffer (Extended Data Fig. 5a). In contrast, neither D-TLKIVW nor D-TLKIVWP exhibited aggregation nor disassembled AD-tau fibrils. These observations suggest that the ability of D-TLKIVWX to fibrilize aids the disassembly of AD-tau fibrils.\u003c/p\u003e\n\u003cp\u003eTo test the hypothesis that D-TLKIVWX form amyloid-like fibrils that disassemble AD-tau, we designed a negative control experiment by eliminating the ability of D-TLKIVWX to fibrilize. As shown in Fig. 2b and Extended Data Fig. 5b,c, 10 mM D-TLKIVWX (X = I, S and R) form well-defined fibrils when they are left undisturbed for three days at room temperature. The corresponding X-ray diffraction analysis (Fig. 2c and insets in Extended Data Fig. 5b,c) confirm these D-TLKIVWX (X = I, S and R) fibrils exhibit the characteristic features of amyloid fibrils with a strong 4.7-4.8 \u0026Aring; reflection corresponding to the distance between hydrogen-bonded \u0026beta;-strands, and a more diffuse 8-12 \u0026Aring; ring arising from the inter-sheet spacing\u003csup\u003e21\u003c/sup\u003e. As such, the aggregation activity of D-TLKIVWX might be prevented by eliminating hydrogen bonds between neighboring \u0026beta;-strands through N-methylation of peptide backbones\u003csup\u003e22\u003c/sup\u003e. Indeed, when we N-methylated the D-isoleucine of D-TLKIVWX (named as D-TLK(N-Me-I)VWX (X = I, S and R)), the peptides were unbale to form fibrils, as shown in Fig. 2d and Extended Data Fig. 5d,e. TEM and dot blot experiments further showed that these non-self-aggregating peptides D-TLK(N-Me-I)VWX (X = I, S and R) were unable to disassemble AD-tau fibrils (Fig. 2e-g and Extended Data Fig. 5f,g). Taken together, these experiments demonstrate the critical role of amyloid-like characteristics of D-TLKIVWX in disassembling AD-tau fibrils.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD-TLKIVWX peptides form right-handed fibrils with conserved motifs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe determined the structures of D-TLKIVWX amyloid-like fibrils, aiming to reveal features that facilitate disassembly of AD-tau fibrils. Generally, \u0026beta;-sheets formed by L-peptides adopt left-handed helical structures, while \u0026beta;-sheets formed by D-peptides adopt right-handed helical structures\u003csup\u003e23,24\u003c/sup\u003e. Here, atomic force microscopy (AFM) confirmed the twist is right-handed in all 18 polymorphs of D-TLKIVWI fibrils observed\u003csup\u003e25,26\u003c/sup\u003e,\u003csup\u003e27\u003c/sup\u003e (Fig. 3a and Extended Data Fig. 6a). Using cryo-EM, we were able to determine the structures of the predominant polymorphs of D-TLKIVWX (X = I, S and R) fibrils (indicated by red squares in Extended Data Fig. 6b-d) at 3.6 \u0026Aring;, 3.5 \u0026Aring;, 3.7 \u0026Aring; resolution, respectively (Fig. 3b-d). Data collection and refinement statistics are summarized in Supplementary Table 1.\u003c/p\u003e\n\u003cp\u003eThe D-TLKIVWX (X = I, S and R) fibrils are each composed of different numbers of protofilaments and these protofilaments associate in different patterns, but all the protofilaments share the same underlying structural motif known as a \u0026quot;steric zipper\u0026quot;\u003csup\u003e28\u003c/sup\u003e, a pair of \u0026beta;-sheets mated together by an interface of snugly fitting sidechains (Fig. 3b-d and Extended Data Fig. 7a,b). In addition, the steric zippers formed by D-TLKIVWX (X = I, S and R) all share the same symmetry pattern in which antiparallel \u0026beta;-sheets mate together by interfacing sidechains of Leu2, Ile4, and Trp6 (an example of \u0026ldquo;class 5\u0026rdquo; symmetry\u003csup\u003e \u003c/sup\u003e\u003csup\u003e29\u003c/sup\u003e). As a result, the helical rise of D-TLKIVWX (X = I, S and R) amyloid-like fibrils is 9.56 \u0026Aring;, instead of the 4.80\u0026thinsp;\u0026Aring; spacing that is common among pathogenic amyloid fibrils. Notably, the sidechain of the seventh residue faces outward from the steric zipper interface in all cases. The identity of the seventh sidechain appears to affect the geometry of association between protofilaments but does not affect the symmetry of the protofilament itself. Thus, the ability to disassemble AD-tau fibrils seems linked more strongly to the steric zipper symmetry (which is conserved among all three functional D-peptides), rather than the pattern of association between zippers (which differs among the three D-peptides). \u003c/p\u003e\n\n\u003cp\u003eImportantly, the structures suggest why the absence of a seventh residue in D-TLKIVW prohibits its fibril formation, and therefore presumably its inability to disassemble tau. Removal of the seventh residue would destabilize fibril formation by reducing the number of backbone hydrogen bonds (10 vs. 14) and increasing the distance between the negative charge of the C-terminal carboxylate and the compensating positive charge at the N-terminal amine of the adjacent strand (5.0 \u0026Aring; vs. 2.8 \u0026Aring;) (Extended Data Fig. 7c). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD-TLKIVWX protofilaments grow along left-handed AD-tau fibrils\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo elucidate how D-TLKIVWX might interact with AD-tau fibrils to induce their disassembly, we cryogenically trapped complexes of D-TLKIVWX (X =I, S and R) with AD-tau fibrils at an intermediate 24-h time point with a lower ratio of D-TLKIVWX to tau (estimated 100:1) in comparison with the time-course experiment in Fig. 2a (500 :1). As a negative control, we also collected images of AD-tau fibrils in the absence of D-TLKIVWX. Helical reconstructions of the AD-tau complexed with each of the three D-peptides revealed the paired helical filament (PHF) tau polymorph\u003csup\u003e30\u003c/sup\u003e (Extended Data Fig. 8), and the atoms modeled into the PHF density showed no significant structural deviations from the negative control (Supplementary Table 2). However, the cryo-EM map of PHF complexed with D-TLKIVWX (X =I, S and R) revealed residual density near Val313-Thr319 of PHF, which was absent from our control (Fig. 4a-c and Extended Data Fig. 9a,b). We attribute the residual density to D-TLKIVWX for two reasons: (1) the shape of the residual density resembles one steric zipper unit of D-TLKIVWX fibrils (Fig. 4a-c and Fig. 3b-d); (2) \u003csup\u003e1\u003c/sup\u003eH-\u003csup\u003e15\u003c/sup\u003eN-HSQC NMR experiments indicated that D-TLKIVWX (X =I and S) interacted with \u003csup\u003e15\u003c/sup\u003eN,\u003csup\u003e13\u003c/sup\u003eC-labeled tau K18+ monomer\u003csup\u003e31\u003c/sup\u003e with chemical shift perturbation mapped to Val306-Lys311\u003csup\u003e11\u003c/sup\u003e and Val313-Thr319 (Fig. 4d and Extended Data Fig. 9c,d), consistent with the location of additional density next to tau PHF. Note that a rapid precipitation of D-TLKIVWX occurred upon mixing in our NMR experiments, corresponding to initial D-peptide fibril formation, but chemical shift changes of monomeric tau were observed over time when soluble fraction of D-peptides increased. Refinement of the 3D reconstruction of D-TLKIVWX (X =I, S and R) complexed with Tau PHF achieved overall resolutions of 3.1 \u0026Aring;, 3.1 \u0026Aring; and 3.5 \u0026Aring;, respectively. We note that the cryo-EM density map of D-TLKIVWI is stronger than that of D-TLKIVWS/R and that D-TLKIVWR is slightly further from the core of tau PHF (Fig. 4a-c and Extended Data Fig. 9b). This difference in occupancy and positioning may explain their unequal efficiency in disassembling AD-tau fibrils (Fig. 1b).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cstrong\u003eStrain-relief of D-TLKIVWX fibrils drives disassembly of AD-tau fibrils\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur observations lead us to propose the strain-relief mechanism of amyloid fibril disassembly illustrated in Fig. 5. As D-TLKIVWX binds to the Tau PHF, it stacks to form D-TLKIVWX amyloid-like protofilaments and these protofilaments are constrained by binding to adopt the left-handed helical twist of Tau PHF (Fig. 5a,b). Because D-TLKIVWX fibrils have an intrinsic right-handed twist (Fig. 3), the cryo-trapped structure of Figure 4a-c is metastable, and a strain develops. If not trapped by freezing, this metastable structure disassembles in hours.\u003c/p\u003e\n\u003cp\u003eWe propose that the strain produced by the metastable left-hand twist of D-TLKIVWX is relieved by the protofilament reversing its twist from a left- to right-handed helix (Fig. 5c). Because the D-peptide fibrils is bound to Tau PHF fibril, the reversal pulls against the Tau PHF structure. As a result, the tau residues that contact D-TLKIVWX are torn away from the native PHF contacts, thereby breaking backbone hydrogen bonds between tau molecules, and permitting solvent to invade the fibril core and further dissociate tau PHF (Fig. 5d). The disassembly products are non-seeding species (Fig. 1); they are tau-D-TLKIVWX complexes, not tau monomers (Extended Data Fig. 4c). This dynamic process is visually depicted in the Supplementary Video 1.\u003c/p\u003e\n\u003cp\u003eOur proposed strain-relief mechanism is consistent with all our experimental findings. First, D-TLKIVWX disassembles tau fibrils to which its precursor, D-TLKIVW, was designed to bind, but does not disassemble other left-handed amyloid fibrils, such as \u0026alpha;-synuclein and wildtype hnRNPA2 LCD fibrils. Second, other analogs of D-TLKIVWX, including D-TLKIVW, D-TLKIVWP, and D-TLK(N-Me-I)VWX (X = I, S, R) fail to disassemble AD-tau fibrils, presumably because they lack the ability to form fibrils themselves. Third, the variation in efficacy of disassembly AD-tau fibrils among D-TLKIVWX can be influenced by their binding strength with AD-tau fibrils. Fourth, the concurrent disappearance of AD-tau fibrils and the newly formed D-TLKIVWX fibrils, along with the emergence of amorphous products at later time points observed in the time-course EM images (Fig. 2a) strongly suggests that the AD-tau fibrils are disassembled.\u003c/p\u003e\n\u003cp\u003eAn additional observation that supports this \u0026ldquo;strain-relief\u0026rdquo; mechanism of disassembly is that L-TLKIVWX (X = C, I, S, R) all display inferior efficacy in disassembling AD-tau fibrils compared to their enantiomers D-TLKIVWX (Extended Data Fig. 10a,b). Despite exhibiting the same fibril-forming property as D-TLKIVWX\u003csup\u003e32\u003c/sup\u003e (Extended Data Fig. 10c) and nearly identical binding to tau monomers (Extended Data Fig. 10d,e), the L-TLKIVWI fibrils principally possess a left-handed twist, same as tau PHF. Therefore, the L-TLKIVWX would have less structural torsion to release when they bind and assemble along the axis of tau PHF in comparison with D-TLKIVWX, resulting in their inferior efficacy in disassembling AD-tau fibrils.\u003c/p\u003e\n\u003cp\u003eThe strain-relief mechanism may be a general theme of action of disruptors of amyloid fibrils. In previous work, we presented evidence that the polyphenolic compound EGCG disassembles AD-tau fibrils by stacking into a metastable amyloid-like fibril on the surface of AD-tau fibrils\u003csup\u003e9\u003c/sup\u003e. A subsequent change in which aromatic rings of EGCG curve into a more stable conformation could provide the energy to disassemble stable AD-tau fibrils. Thus, both disassembling actions of the very different compounds EGCG and D-TLKIVWX on AD-tau fibrils may be considered examples of strain-relief mechanisms.\u003c/p\u003e\n\u003cp\u003eIn summary, we find the disassembly of AD-tau fibrils is not exclusive to D-TLKIVWC, because D-TLKIVWX (X = A, S, D, I, V, R, K, E, T) also displays this property, but with unequal efficacy. We find that the amyloid-like, fibril-forming property of D-TLKIVWX contributes to the disassembly of AD-tau fibrils. Based on cryo-EM, AFM, NMR, and other results reported here, we propose that the disassembly of AD-tau fibrils is driven by the release of torsion in D-TLKIVWX protofilaments. It remains unknown whether D-TLKIVWX disassembles tau fibrils from other tauopathies\u003csup\u003e33\u003c/sup\u003e. The strain-relief mechanism of amyloid disassembly proposed here may explain how diverse small molecules can provide sufficient energy to disrupt extremely stable pathogenic amyloid fibrils. This mechanism may be applied to the design of a new generation of disaggregators for tau and other pathological amyloids.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eChemicals and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll the peptides were synthesized by GenScript and purified to ≥98%, as determined by mass spectrometry and HPLC (GenScript Corp, Piscataway, NJ).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRecombinant protein expression and purification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnlabeled recombinant tau K18+ (residues Gln244-Glu380 of 4R tau) was expressed in a pNG2 vector in BL21-Gold \u003cem\u003eE. coli\u003c/em\u003e cells grown in LB to an A\u003csub\u003e600\u003c/sub\u003e = 0.8. Cells were induced with 0.5 mM isopropyl 1-thio-β-D-galactopyranoside (IPTG) for 3 h at 37 °C and lysed by sonication in 20 mM MES buffer (pH 6.8) with 1 mM EDTA, 1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 1 mM dithiothreitol (DTT), and HALT protease inhibitor before the addition of NaCl (500 mM final concentration). The lysate was boiled for 20 min and then clarified by centrifugation at 15,000 rpm for 15 min and dialyzed to 20 mM MES buffer (pH 6.8) with 50 mM NaCl and 5 mM DTT. Dialyzed lysate was purified on a 5-mL HiTrap SP HP ion exchange column and eluted over a gradient of NaCl from 50 to 550 mM. Protein was further purified on a HiLoad 16/600 Superdex 75 pg column (GE Healthcare) in 10 mM Tris (pH 7.6) with 100 mM NaCl and 1 mM DTT and concentrated to 20-60 mg/mL by ultrafiltration using a 3-kDa cutoff filter (Millipore-Sigma, Burlington, MA).\u003c/p\u003e\n\u003cp\u003eIsotopically labeled tau K18+ proteins for NMR were grown in M9 H\u003csub\u003e2\u003c/sub\u003eO media supplemented with \u003csup\u003e15\u003c/sup\u003eNH\u003csub\u003e4\u003c/sub\u003eCl (and \u003csup\u003e13\u003c/sup\u003eC-glucose) as the sole nitrogen (and carbon) source. Protein expression was induced with 1 mM IPTG at 37 °C for 6 hours. The purification was same as for the unlabeled protein.\u003c/p\u003e\n\u003cp\u003eThe construct for overexpression of mCherry-hnRNPA2-LCD fusion protein was provided by Dr. Masato Kato of University of Texas, Southwestern. Protein overexpression and purification procedures followed the same protocol reported previously\u003csup\u003e34\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExtraction of tau fibrils from AD patient brains\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFrozen brain tissues were weighed, diced into small pieces, and resuspended in 10 mL/gram of sucrose buffer (800 mM NaCl, 10% sucrose, 10 mM Tris-HCl, pH 7.4, 0.1 mM EDTA, 1 mM DTT) supplemented with 1:100 (v/v) Halt protease inhibitor (Thermo Scientific). Resuspended tissue was homogenized using a Polytron homogenizer (Thomas Scientific) and centrifuged at 20,000 × g at 4 °C for 20 minutes. The crude supernatant was treated with N-lauroylsarcosinate (1% [w/v] final concentration) and shaken at room temperature (22 °C) for 1 h. The supernatant was then centrifuged at 100,000 × g for 1 h at 4 °C. The sarkosyl-insoluble pellet was resuspended in washing buffer (10 mM Tris–HCl, pH 7.4, 800 mM NaCl, 5 mM EDTA, 1 mM EGTA, 1 mM DTT, 10% sucrose) and centrifuged at 20,100 x g for 30 mins at 4 °C. After centrifugation, the supernatant was further centrifuged at 100,000 × g (Beckman Coulter, Optima MAX-XP) for 1 h at 4 °C. Finally, the purified sarkosyl-insoluble pellet was resuspended in 250 μl 20 mM Tris-HCl, pH 7.4, 100 mM NaCl and stored at -80 °C.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThioflavin T (ThT) assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKinetic fluorescence data were collected in a microplate reader (FLUOstar Omega, BMG Labtech) at 37 ˚C with double orbital shaking at 700 rpm. Fluorescence measurements were recorded every 10 mins with excitation and emission wavelengths of 440 and 480 nm. All samples were added in triplicate and experiments were repeated at least twice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDot blot assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePurified AD-tau fibrils from brain extract were incubated with 500 μM L/D-TLKIVW, D-TLKIVWX (X = C, A, S, D, I, V, R, K, E, P, T), D-TLK(N-Me-I)VWX (X =I, S and R), L-TLKIVWX (X = C, I, S and R) at 37 ℃ for 48 h, respectively. 2.5 μL of samples were added on nitrocellulose membrane (0.2 µm, Bio-Rad). The membrane was blocked by 5% (w/v) nonfat dry milk in TBS-T (T = 0.1% (v/v) Tween-20) at room temperature for 1 hr. After blocking, the membrane was incubated with GT38 antibody obtained from Virginia Lee’s lab (1:1000) in 5% (w/v) milk in TBS-T at 4 ℃ overnight. Then, the membrane was washed in TBS-T three times for 5 minutes each and incubated with goat anti-mouse IgG HRP (1:5000, cat# AB205719, Abcam) in TBS-T for 1 h at room temperature. The membrane was washed three more times, and the signal was developed with Pierce\u003csup\u003eTM\u003c/sup\u003e ECL western blotting substrate (170-5061, BioRad).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNegative stain transmission electron microscopy (TEM)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e6 μL of sample was applied to a glow discharged carbon coated electron microscopy grid (CF150-Cu, Electron Microscopy Sciences) for 5 minutes. Then grids were stained with 2% uranyl acetate for 2 minutes. Samples were visualized using a FEI Tecnai T12 Quick room temperature transmission electron microscope equipped with a Gatan 2,048 x 2,048 CCD camera operated at an acceleration voltage of 120 kV.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAD brain tau fibril seeding in tau biosensor cell line\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHEK293 cell lines stably expressing tau-K18-YFP were engineered by Marc Diamond’s laboratory at the University of Texas Southwestern Medical Center and used without further characterization or authentication. Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies, cat# 11965092) supplemented with 10% (v/v) fetal bovine serum (Life Technologies, cat# A3160401), 1% antibiotic-antimycotic (Life Technologies, cat# 15140122), and 1% Glutamax (Life Technologies, cat# 35050061) at 37 °C, 5% CO\u003csub\u003e2\u003c/sub\u003e in a humidified incubator. AD-tau fibrils were incubated with D-TLKIVWX (X = A, S, D, I, V, R, K, E, P, T) (2.5, 5, 10, 20, 50, 75, 100 μM) at 4 °C overnight and sonicated in a cup horn water bath for 3 min. Then these disassembly products of AD-tau were mixed with 1 volume of Lipofectamine 3000 prepared by diluting 1μL of Lipofectamine 3000 (Life Technologies, cat# 2729899) in 19 μL Opti-MEM (Life Technologies, cat# 31985070). After 20 min, 10 μL of fibrils were added to 90 μL tau biosensor cells. After 24 hours of incubation, the number of seeded aggregates was determined by imaging the entire well of a 96-well plate in triplicate using a Celigo image cytometer (Nexcelom) in the YFP channel. The data analysis was described before. For high-quality images, cells were photographed on a ZEISS Axio Observer D1 fluorescence microscope using the EGFP fluorescence channel.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell toxicity of AD-tau disassembly species\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAD-tau fibril disassembly species were produced by incubating AD-tau fibrils (estimated 1 μM) with D-TLKIVWX (5, 10, 20, 50, or 100 μM) at 4 °C overnight. Neuro 2A (N2a) cells were cultured in DMEM supplemented with 10% (v/v) fetal bovine serum, 1% antibiotic-antimycotic, and 1% Glutamax in a 5% CO\u003csub\u003e2\u003c/sub\u003e humidified environment at 37 °C. Cells were plated at a density of roughly 6,000 cells/well on 96-well plates in 90 μL of fresh medium. After 24 h, 10 μL of the above AD-tau fibril disassembly species were added and the cells were incubated for another 24 h at 37 °C. Cytotoxicity was measured utilizing an MTT assay.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWestern blot\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePurified brain-extracted AD-tau fibrils were incubated with 500 μM D-TLKIVWI at 37 ℃ for 48 h. The sample was centrifuged at 21,000 x g for 30 min at 4 ℃ (Eppendorf Centrifuge 5424R). Western blot was performed with anti-tau rabbit polyclonal antibody (1:1000, Dako A0024) and anti-rabbit HRP-conjugated secondary antibody (1:5000, Thermo Fisher Scientific).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eX-ray diffraction (XRD)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eD-TLKIVWX (X = I, S, R) peptides were dissolved to 10 mM in deionized water and incubated at room temperature quiescently for three days. Peptide fibrils were aligned by pipetting the suspension in a 3 mm gap between two fire-polished glass rods and drying overnight. The aligned fibrils were cooled to 100 K. Diffraction data was collected on a FR-E+ rotating anode x-ray generator (Rigaku) equipped with a R-AXIS HTC imaging plate detector (Rigaku). Cu K-α x-ray beam with 1.5406 Å wavelength was used, and the detector was placed 78 mm from the sample. Diffraction images were visualized using ADXV (The Scripps Research Institute).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAtomic force microscopy (AFM)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e4 mM D-TLKIVWI in deionized water was shaken at room temperature for three days and diluted into distilled water in a 1:10 ratio. Then, 5 μL of diluted sample was deposited onto freshly cleaved mica and incubated for 10 min. The sample was rinsed with Milli-Q water and dried under a stream of nitrogen gas. AFM images were collected using a Dimension Icon microscope (Bruker) in PeakForce Tapping mode using ScanAsyst-HR probes. Each collected image had a scan size of 3 x 3 μm and 2048 x 2048 pixels and was collected using a scan rate of 0.494 Hz. Nanoscope Analysis software (Version 2.0, Bruker) was used to process the image data by flattening the height topology data to remove tilt and scanner bow. Fibrils were traced and computationally straightened from collected AFM images in Matlab using Trace_y\u003csup\u003e35\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCryo-EM samples\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eD-TLKIVWI fibrils were optimized by shaking 4 mM D-TLKIVWI in deionized water at room temperature for three days. 10 mM D-TLKIVWS/R in deionized water formed fibrils when left undisturbed for three days at room temperature. For D-TLKIVWR fibrils, the pH of the peptide solution was adjusted to 7.0. Prior to cryo-EM grid preparation, AD-tau fibrils in a buffer comprised of 20 mM Tris-HCl pH 7.4, 100 mM NaCl were pre-incubated at 37 °C with final concentration of 100 µM D-TLKIVWX (X = I, S, R) from 10 mM stocking solution in water for 24 hours. Control tau fibrils from the same brain donor were treated identically except for the addition of D-TLKIVWX.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCryo-EM data collection and processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo prepare the cryo-EM grids, we applied 2.5 μl of sample solution onto Quantifoil 1.2/1.3 200 mesh electron microscope grids glow-discharged for 2 minutes in a Pelco easiGlow unit before use. Grids were plunge-frozen into liquid nitrogen-cooled liquid ethane inside a Vitrobot Mark IV (FEI) vitrification robot after blotting. Cryo-EM data of D-TLKIVWR and D-TLKIVWS fibrils were collected on a Titan Krios transmission electron microscope (Thermo Fisher Scientific) located at the National Center for Cryo-EM Access and Training, which is equipped with a Bioquantum/K3 direct detection camera (Gatan), operated with 300 kV acceleration voltage and an energy slit width of 20 eV, automated with Leginon software package\u003csup\u003e36\u003c/sup\u003e. Super-resolution movies were collected with a calibrated pixel size of 1.067 Å/pixel (0.5335 Å/pixel in super-resolution movie frames) and a dose per frame of ~1.5 e-/Å\u003csup\u003e2\u003c/sup\u003e. A total of 40 frames with a frame rate of 12 Hz were taken for each movie, resulting in a final dose of ~60 e-/Å\u003csup\u003e2\u003c/sup\u003e per image. D-TLKIVWI fibrils were collected on a Titan Krios located at the HHMI Janelia Research Campus, which is equipped with a cold-FEG source (CFEG), a Selectris X energy filter and a Falcon 4i direct detection camera (TFS), operated with 300 kV acceleration voltage and an energy slit width of 6 eV, and automated with the SerialEM software package\u003csup\u003e37\u003c/sup\u003e. Electron Event Representation (EER) files were collected with a calibrated pixel size of 0.94 Å/pixel and a dose per raw frame of 0.0244 e-/Å\u003csup\u003e2\u003c/sup\u003e, resulting in 55 e-/Å\u003csup\u003e2\u003c/sup\u003e per image. AD-tau/D-TLKIVWX (X = I, S and R) were collected similarly as D-TLKIVWI fibrils, although manually targeted in SerialEM package. The AD-tau control was collected on a Titan Krios/Bioquantum/K3 setup located at the Stanford-SLAC Cryo-EM Center, operated with 300 kV acceleration voltage and an energy slit width of 20 eV, automated with EPU (TFS). (See details in Supplementary Table 1,2).\u003c/p\u003e\n\u003cp\u003eMovies and EER files were motion-corrected in RELION\u003csup\u003e38\u003c/sup\u003e and binned to pixel sizes according to Supplementary Table 1,2. CTF estimation was performed using CTFFIND4\u003csup\u003e39\u003c/sup\u003e. AD-tau/D-TLKIVWX fibrils were manually picked using e2helixboxer.py from EMAN2\u003csup\u003e40\u003c/sup\u003e. D-TLKIVWX fibrils and AD-tau control particle picking was initially done manually using e2helixboxer.py from EMAN2 for about 100 images as a training set for crYOLO\u003csup\u003e41\u003c/sup\u003e. CrYOLO was then trained with default parameters and was used to pick the rest of the images. Particle extraction, two-dimensional classification, three-dimensional classification, and 3D refinement were performed in RELION\u003csup\u003e42\u003c/sup\u003e. Briefly, particles were initially extracted using a larger box size of 640 pixels with two-fold binning. 2D classification was then performed with all particles to eliminate bad particles and group particles into polymorphs if necessary. Particles from each polymorph were selected, extracted with smaller box sizes at native pixel sizes of detectors (binning=1), further “purified” using 2D classification and subjected to 3D classification, which was done initially with one class and then with three classes, using a Gaussian cylinder as the initial model. The best 3D classes were selected, and corresponding particles were finally refined with 3D auto-refine for the reported maps. (See details in Table Supplementary Table 1,2). Part of the Cryo-EM data processing used Expanse GPU at San Diego Supercomputer Center through allocation BIO230174 from the Advanced Cyberinfrastructure Coordination Ecosystem\u003csup\u003e43\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAtomic model building\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur starting atomic model of D-TLKIVWI was an ideal β-strand. It was manually adjusted to fit the electrostatic potential map using Coot\u003csup\u003e44\u003c/sup\u003e and automated refinement was performed using Phenix\u003csup\u003e45\u003c/sup\u003e. To facilitate good rotamer geometry, the initial building and refinement was performed using a map with handedness chosen so that the amino acid residues appeared to be levorotary rather than dextrorotary. In this way, we could take advantage of the rotamer library in Coot which exists for L-amino acids, but not for D-amino acids. In the final step of refinement, the map and coordinates were inverted to the correct hand, consistent with D-amino acids. The starting models for D-TLKIVWS and D-TLKIVWR were adapted from the refined D-TLKIVWI structure. All atomic models were refined in successive rounds using Coot for manual building and Phenix for automated refinement. Model validation statistics of all three D-peptide structures are reported in Supplementary Table 1.\u003c/p\u003e\n\u003cp\u003eOur starting atomic model of the complex between AD-tau PHF and D-TLKIVWI was built by manually orienting coordinates of the tau PHF (PDB ID 7nrv)\u003csup\u003e46\u003c/sup\u003e to fit the electrostatic potential map using Coot\u003csup\u003e44\u003c/sup\u003e and then refined with Phenix. Coordinates of a pair of β-sheets were extracted from the D-TLKIVWI structure described above and manually docked on the surface of the PHF using guidance from the 3.1 Å cryoEM map, as well as the low-pass filtered map (7 Å). We noted a blob of residual density situated at the end of three lysine side chains: K317 and K321 of tau and K3 of the D-peptide. Whatever molecule produced this residual density does not depend on the presence of the D-peptide to bind to tau, since a similar blob was evident in our PHF negative control lacking D-peptide. Indeed, the presence of this residual density was noted in the original structure report of AD-tau PHFs\u003csup\u003e30\u003c/sup\u003e, and even noted in maps from PHFs produced with recombinant tau (PDB ID 7ql4)\u003csup\u003e47\u003c/sup\u003e. The chemical environment of this blob suggests that the blob originates from an anion, but the density is not sufficiently detailed to uniquely identify the chemical species. It is roughly the size of a pair of phosphate ions. We chose to model this residual density with ethylenediaminetetraacetate (EDTA) because it fits the density, caries the expected negative charges to complement the positive charge on K317 and K321, and we know that EDTA was included in the buffer used for PHF purification. The starting models for tau complexed with D-TLKIVWS and D-TLKIVWR were obtained using an analogous procedure. The final refined coordinates for these two complexes do not include the D-peptides because density for the peptides was visible only in the low-pass filtered maps, and not in the high-resolution map (Figure 4 and Extended data Figure 9a).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNMR spectroscopy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNMR samples were ~0.5 mL of 0.1 mM \u003csup\u003e15\u003c/sup\u003eN,\u003csup\u003e13\u003c/sup\u003eC-labeled tau K18+ protein in 100 mM KCl, 20 mM NaH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, 1 mM TCEP, 5%/95% D\u003csub\u003e2\u003c/sub\u003eO/H\u003csub\u003e2\u003c/sub\u003eO, pH 7.0 without or with 1 mM D-TLKIVWI, D-TLKIVWS, L-TLKIVWI. All NMR spectra were acquired at 25 °C with Bruker Avance III HD 600 MHz spectrometer equipped with QCI HCNP cryoprobe or Avance Neo 800 MHz spectrometer equipped with TCI HCN cryoprobe. Backbone assignments for both free tau K18+ and D-TLKIVWI bound tau K18+ were carried out using HNCACB, CBCA(CO)NH and C(CO)NH NMR experiments. NMR spectra were acquired with Topspin (Bruker), processed with NMRPipe\u003csup\u003e48\u003c/sup\u003e, and analyzed with NMRFAM-Sparky\u003csup\u003e49\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFor the chemical shift perturbation (CSP) analysis, the overall change in chemical shift Δ was calculated between the free and bound states of tau K18+ protein as\u003csup\u003e50\u003c/sup\u003e:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"256\" height=\"51\"\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003ewhere Δδ\u003csub\u003eH\u003c/sub\u003e and Δδ\u003csub\u003eN\u003c/sub\u003e are the differences between the \u003csup\u003e1\u003c/sup\u003eH\u003csub\u003eN\u003c/sub\u003e and \u003csup\u003e15\u003c/sup\u003eN chemical shifts of the two states being compared.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGraphs are expressed as means + standard deviation (SD) and data were analyzed using SPSS 25 statistical analysis software (SPSS, Chicago, IL, USA). The one-way analysis of variance (ANOVA) was used to analyze difference among multiple groups. Statistical differences for all tests were considered significant at the *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001 levels, NS, not significant.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eA\u0026beta;, amyloid beta protein; AD, Alzheimer\u0026rsquo;s disease, AD-tau, tau fibrils extracted from AD brains; All D-peptides TLKIVWX (D-TLKIVWX), in which X can be any enantiomorph of the 20 coded amino acid residues; PHFs, paired helical filaments, the most abundant polymorph of AD-tau fibril.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eReporting summary\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFurther information on research design is available in the Nature Research Reporting Summary linked to this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCryo-EM maps and atomic models of the fibrils from peptides D-TLKIVWX (X = I, S, R) alone have been deposited in the Electron Microscopy Data Bank (EMDB) and PDB under the accession codes EMD-44181/9B4I (D-TLKIVWI), EMD-44182/9B4J (D-TLKIVWS), and EMD-44183/9B4K (D-TLKIVWR). Cryo-EM maps and atomic models of paired helical filament of Tau from Alzheimer\u0026rsquo;s patient in complex with the three peptides have been deposited in the same databases under the accession codes EMD-44184/9B4L (Tau-D-TLKIVWI), EMD-44185/9B4M (Tau-D-TLKIVWS), and EMD-44186/9B4N (Tau-D-TLKIVWR). Cryo-EM map and atomic model of the paired helical filament from the same patient have been deposited under the accession codes EMD-44187/9B4O. All data presented in this article are available within the figures and Supplementary Information files. All other data are available from the corresponding authors upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgment\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the donors and their families for the AD brain tissue; without whom this work would not have been possible. We acknowledge NIH 1R01AG070895 (D.S.E.), NIH RF1AG065407 (D.S.E.), DOE-FC02-02ERG and Alzheimer\u0026apos;s Association Research Fellowship AARF-21-848751 (K.H.) for support. The authors thank Dr. Marc Diamond for sharing the YFP-labeled tau K18 expressing HEK293 cells. We thank Virginia Lee for generously gifting the GT38 antibody. Some of this work was performed at the National Center for CryoEM Access and Training (NCCAT) and the Simons Electron Microscopy Center located at the New York Structural Biology Center, supported by the NIH Common Fund Transformative High Resolution Cryo-Electron Microscopy program (U24 GM129539), and by grants from the Simons Foundation (SF349247) and NY State Assembly. We also thank the staff at the HHMI Janelia CryoEM Facility for help and support. Some of this work was performed at the Stanford-SLAC Cryo-EM Center (S2C2) supported by the NIH Common Fund Transformative High Resolution Cryo-Electron Microscopy program (U24 GM129541). This work used Expanse GPU at San Diego Supercomputer Center through allocation BIO230174 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services \u0026amp; Support (ACCESS) program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296. We acknowledge support of NMR equipment by grants from NIH (S10OD016336 and S10OD025073) and from DOE (DE-FC02-02ER63421). Y.Y. was supported in part by R35GM131901 to J.F.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe project was conceived and designed by K.H. and D.S.E. Cell seeding experiment was performed by J.L.D. and K.H.. Dot blot, TEM and western blot were performed by K.H. and J.P.. J.Z. and J.L.D. provided the recombinant tau K18+. J.P. extracted tau fibrils from AD patient brains with the help of X.C.. X-ray diffraction was done by K.H. and Y.X.J.. AFM was carried out by L.L.. Cryo-EM grids were prepared by K.H., Y.X.J. and P.G.. Cryo-EM data were collected by P.G. with assistance from S.Y., Z.Y., Y.X.J. and D.R.B.. K.H. and P.G. processed cryo-EM data with assistance from Y.X.J., D.R.B., X.C. and J.L.. M.R.S. and K.H. built atomic models. Isotopically labeled tau K18+ was purified by K.H. NMR experiments were conducted and analyzed by Y.Y. and J.F.. The manuscript was prepared by K.H., M.R.S., P.G., and D.S.E. with contributions from all the other authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eD.S.E. is SAB chair and equity holder of ADRx, Inc. All other authors declare no conflicts. Part of the work was disclosed in our provisional patent application (Serial No. 63/510,194).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eArriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T. \u0026amp; Hyman, B. T. Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer\u0026rsquo;s disease. \u003cem\u003eNeurology\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 631\u0026ndash;631 (1992).\u003c/li\u003e\n\u003cli\u003eKollmer, M. \u003cem\u003eet al.\u003c/em\u003e Cryo-EM structure and polymorphism of A\u0026beta; amyloid fibrils purified from Alzheimer\u0026rsquo;s brain tissue. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 4760 (2019).\u003c/li\u003e\n\u003cli\u003eScheres, S. H. W., Ryskeldi-Falcon, B. \u0026amp; Goedert, M. Molecular pathology of neurodegenerative diseases by cryo-EM of amyloids. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e621\u003c/strong\u003e, 701\u0026ndash;710 (2023).\u003c/li\u003e\n\u003cli\u003eLow, K. J. Y., Venkatraman, A., Mehta, J. S. \u0026amp; Pervushin, K. Molecular mechanisms of amyloid disaggregation. \u003cem\u003eJ. Adv. Res.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 113\u0026ndash;132 (2022).\u003c/li\u003e\n\u003cli\u003eIadanza, M. G., Jackson, M. P., Hewitt, E. W., Ranson, N. A. \u0026amp; Radford, S. E. A new era for understanding amyloid structures and disease. \u003cem\u003eNat. Rev. Mol. Cell Biol.\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 755\u0026ndash;773 (2018).\u003c/li\u003e\n\u003cli\u003eNachman, E. \u003cem\u003eet al.\u003c/em\u003e Disassembly of Tau fibrils by the human Hsp70 disaggregation machinery generates small seeding-competent species. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e295\u003c/strong\u003e, 9676\u0026ndash;9690 (2020).\u003c/li\u003e\n\u003cli\u003eAl-Hilaly, Y. K. \u003cem\u003eet al.\u003c/em\u003e Cysteine-Independent Inhibition of Alzheimer\u0026rsquo;s Disease-like Paired Helical Filament Assembly by Leuco-Methylthioninium (LMT). \u003cem\u003eJ. Mol. Biol.\u003c/em\u003e \u003cstrong\u003e430\u003c/strong\u003e, 4119\u0026ndash;4131 (2018).\u003c/li\u003e\n\u003cli\u003eLow, K. J. Y., Venkatraman, A., Mehta, J. S. \u0026amp; Pervushin, K. Molecular mechanisms of amyloid disaggregation. \u003cem\u003eJ. Adv. Res.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 113\u0026ndash;132 (2022).\u003c/li\u003e\n\u003cli\u003eSeidler, P. M. \u003cem\u003eet al.\u003c/em\u003e Structure-based discovery of small molecules that disaggregate Alzheimer\u0026rsquo;s disease tissue derived tau fibrils in vitro. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 5451 (2022).\u003c/li\u003e\n\u003cli\u003eWischik, C. M., Harrington, C. R. \u0026amp; Storey, J. M. Tau-aggregation inhibitor therapy for Alzheimer\u0026rsquo;s disease. \u003cem\u003eBiochem. Pharmacol.\u003c/em\u003e \u003cstrong\u003e88\u003c/strong\u003e, 529\u0026ndash;539 (2014).\u003c/li\u003e\n\u003cli\u003eSievers, S. A. \u003cem\u003eet al.\u003c/em\u003e Structure-based design of non-natural amino-acid inhibitors of amyloid fibril formation. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e475\u003c/strong\u003e, 96\u0026ndash;100 (2011).\u003c/li\u003e\n\u003cli\u003eChong, P., Sia, C., Tripet, B., James, O. \u0026amp; Klein, M. Comparative immunological properties of enantiomeric peptides. \u003cem\u003eLett. Pept. Sci.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 99\u0026ndash;106 (1996).\u003c/li\u003e\n\u003cli\u003ePeters, O., Cosma, N. C., Kutzsche, J. \u0026amp; Willbold, D. A randomized, placebo‐controlled, double‐blind, Phase 1b study to evaluate the safety, tolerability and pharmacodynamics of PRI‐002 in early AD. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, e069253 (2022).\u003c/li\u003e\n\u003cli\u003eHou, K. \u003cem\u003eet al.\u003c/em\u003e D-peptide-magnetic nanoparticles fragment tau fibrils and rescue behavioral deficits in a mouse model of Alzheimer\u0026rsquo;s disease. \u003cem\u003eSci. Adv.\u003c/em\u003e (2024), in press.\u003c/li\u003e\n\u003cli\u003eWeismiller, H. A., Holub, T. J., Krzesinski, B. J. \u0026amp; Margittai, M. A thiol-based intramolecular redox switch in four-repeat tau controls fibril assembly and disassembly. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e297\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003ePrifti, E. \u003cem\u003eet al.\u003c/em\u003e The two cysteines of tau protein are functionally distinct and contribute differentially to its pathogenicity in vivo. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 797\u0026ndash;810 (2021).\u003c/li\u003e\n\u003cli\u003eKumat, T., Samuel, D., Jayaraman, G., Srimathi, T. \u0026amp; Yu, C. The role of proline in the prevention of aggregation during protein folding in vitro. \u003cem\u003eIUBMB Life\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 509\u0026ndash;517 (1998).\u003c/li\u003e\n\u003cli\u003eGibbons, G. S. \u003cem\u003eet al.\u003c/em\u003e Detection of Alzheimer\u0026rsquo;s disease (AD) specific tau pathology with conformation-selective anti-tau monoclonal antibody in co-morbid frontotemporal lobar degeneration-tau (FTLD-tau). \u003cem\u003eActa Neuropathol. Commun.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 1\u0026ndash;12 (2019).\u003c/li\u003e\n\u003cli\u003eWillbold, D., Strodel, B., Schr\u0026ouml;der, G. F., Hoyer, W. \u0026amp; Heise, H. Amyloid-type Protein Aggregation and Prion-like Properties of Amyloids. \u003cem\u003eChem. Rev.\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 8285\u0026ndash;8307 (2021).\u003c/li\u003e\n\u003cli\u003eHolmes, B. B. \u003cem\u003eet al.\u003c/em\u003e Proteopathic tau seeding predicts tauopathy in vivo. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, E4376\u0026ndash;E4385 (2014).\u003c/li\u003e\n\u003cli\u003eSerpell, L. C. Alzheimer\u0026rsquo;s amyloid fibrils: structure and assembly. \u003cem\u003eBiochim. Biophys. Acta BBA-Mol. Basis Dis.\u003c/em\u003e \u003cstrong\u003e1502\u003c/strong\u003e, 16\u0026ndash;30 (2000).\u003c/li\u003e\n\u003cli\u003eHughes, E., Burke, R. M. \u0026amp; Doig, A. J. Inhibition of toxicity in the \u0026beta;-amyloid peptide fragment \u0026beta;-(25\u0026ndash;35) using N-methylated derivatives: a general strategy to prevent amyloid formation. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e275\u003c/strong\u003e, 25109\u0026ndash;25115 (2000).\u003c/li\u003e\n\u003cli\u003eWang, M. \u003cem\u003eet al.\u003c/em\u003e Left or Right: How Does Amino Acid Chirality Affect the Handedness of Nanostructures Self-Assembled from Short Amphiphilic Peptides? \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e139\u003c/strong\u003e, 4185\u0026ndash;4194 (2017).\u003c/li\u003e\n\u003cli\u003eKe, P. C. \u003cem\u003eet al.\u003c/em\u003e Half a century of amyloids: past, present and future. \u003cem\u003eChem. Soc. Rev.\u003c/em\u003e \u003cstrong\u003e49\u003c/strong\u003e, 5473\u0026ndash;5509 (2020).\u003c/li\u003e\n\u003cli\u003eAdamcik, J. \u0026amp; Mezzenga, R. Study of amyloid fibrils via atomic force microscopy. \u003cem\u003eCurr. Opin. Colloid Interface Sci.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 369\u0026ndash;376 (2012).\u003c/li\u003e\n\u003cli\u003eLutter, L. \u003cem\u003eet al.\u003c/em\u003e Structural Identification of Individual Helical Amyloid Filaments by Integration of Cryo-Electron Microscopy-Derived Maps in Comparative Morphometric Atomic Force Microscopy Image Analysis. \u003cem\u003eJ. Mol. Biol.\u003c/em\u003e \u003cstrong\u003e434\u003c/strong\u003e, 167466 (2022).\u003c/li\u003e\n\u003cli\u003eClose, W. \u003cem\u003eet al.\u003c/em\u003e Physical basis of amyloid fibril polymorphism. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 699 (2018).\u003c/li\u003e\n\u003cli\u003eNelson, R. \u003cem\u003eet al.\u003c/em\u003e Structure of the cross-\u0026beta; spine of amyloid-like fibrils. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e435\u003c/strong\u003e, 773\u0026ndash;778 (2005).\u003c/li\u003e\n\u003cli\u003eSawaya, M. R. \u003cem\u003eet al.\u003c/em\u003e Atomic structures of amyloid cross-\u0026beta; spines reveal varied steric zippers. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e447\u003c/strong\u003e, 453\u0026ndash;457 (2007).\u003c/li\u003e\n\u003cli\u003eFitzpatrick, A. W. \u003cem\u003eet al.\u003c/em\u003e Cryo-EM structures of tau filaments from Alzheimer\u0026rsquo;s disease. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e547\u003c/strong\u003e, 185\u0026ndash;190 (2017).\u003c/li\u003e\n\u003cli\u003eBarr\u0026eacute;, P. \u0026amp; Eliezer, D. Structural transitions in tau k18 on micelle binding suggest a hierarchy in the efficacy of individual microtubule‐binding repeats in filament nucleation. \u003cem\u003eProtein Sci.\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 1037\u0026ndash;1048 (2013).\u003c/li\u003e\n\u003cli\u003eCribbs, D. H., Pike, C. J., Weinstein, S. L., Velazquez, P. \u0026amp; Cotman, C. W. All-D-enantiomers of \u0026beta;-amyloid exhibit similar biological properties to all-L-\u0026beta;-amyloids. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e272\u003c/strong\u003e, 7431\u0026ndash;7436 (1997).\u003c/li\u003e\n\u003cli\u003eShi, Y. \u003cem\u003eet al.\u003c/em\u003e Structure-based classification of tauopathies. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e598\u003c/strong\u003e, 359\u0026ndash;363 (2021).\u003c/li\u003e\n\u003cli\u003eLu, J. \u003cem\u003eet al.\u003c/em\u003e CryoEM structure of the low-complexity domain of hnRNPA2 and its conversion to pathogenic amyloid. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 4090 (2020).\u003c/li\u003e\n\u003cli\u003eXue, W.-F. Trace_y: Software algorithms for structural analysis of individual helical filaments by three-dimensional contact point reconstruction atomic force microscopy. \u003cem\u003ebioRxiv\u003c/em\u003e 2023\u0026ndash;07 (2023).\u003c/li\u003e\n\u003cli\u003eSuloway, C. \u003cem\u003eet al.\u003c/em\u003e Automated molecular microscopy: the new Leginon system. \u003cem\u003eJ. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e151\u003c/strong\u003e, 41\u0026ndash;60 (2005).\u003c/li\u003e\n\u003cli\u003eMastronarde, D. N. Automated electron microscope tomography using robust prediction of specimen movements. \u003cem\u003eJ. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e152\u003c/strong\u003e, 36\u0026ndash;51 (2005).\u003c/li\u003e\n\u003cli\u003eScheres, S. H. RELION: implementation of a Bayesian approach to cryo-EM structure determination. \u003cem\u003eJ. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e180\u003c/strong\u003e, 519\u0026ndash;530 (2012).\u003c/li\u003e\n\u003cli\u003eRohou, A. \u0026amp; Grigorieff, N. CTFFIND4: Fast and accurate defocus estimation from electron micrographs. \u003cem\u003eJ. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e192\u003c/strong\u003e, 216\u0026ndash;221 (2015).\u003c/li\u003e\n\u003cli\u003eTang, G. \u003cem\u003eet al.\u003c/em\u003e EMAN2: an extensible image processing suite for electron microscopy. \u003cem\u003eJ. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e157\u003c/strong\u003e, 38\u0026ndash;46 (2007).\u003c/li\u003e\n\u003cli\u003eWagner, T. \u003cem\u003eet al.\u003c/em\u003e Two particle-picking procedures for filamentous proteins: SPHIRE-crYOLO filament mode and SPHIRE-STRIPER. \u003cem\u003eActa Crystallogr. Sect. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e76\u003c/strong\u003e, 613\u0026ndash;620 (2020).\u003c/li\u003e\n\u003cli\u003eHe, S. \u0026amp; Scheres, S. H. Helical reconstruction in RELION. \u003cem\u003eJ. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e198\u003c/strong\u003e, 163\u0026ndash;176 (2017).\u003c/li\u003e\n\u003cli\u003eBoerner, T. J., Deems, S., Furlani, T. R., Knuth, S. L. \u0026amp; Towns, J. Access: Advancing innovation: Nsf\u0026rsquo;s advanced cyberinfrastructure coordination ecosystem: Services \u0026amp; support. in \u003cem\u003ePractice and Experience in Advanced Research Computing\u003c/em\u003e 173\u0026ndash;176 (2023).\u003c/li\u003e\n\u003cli\u003eEmsley, P. \u0026amp; Cowtan, K. Coot: model-building tools for molecular graphics. \u003cem\u003eActa Crystallogr. D Biol. Crystallogr.\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 2126\u0026ndash;2132 (2004).\u003c/li\u003e\n\u003cli\u003eLiebschner, D. \u003cem\u003eet al.\u003c/em\u003e Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. \u003cem\u003eActa Crystallogr. Sect. Struct. Biol.\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 861\u0026ndash;877 (2019).\u003c/li\u003e\n\u003cli\u003eShi, Y. \u003cem\u003eet al.\u003c/em\u003e Cryo-EM structures of tau filaments from Alzheimer\u0026rsquo;s disease with PET ligand APN-1607. \u003cem\u003eActa Neuropathol. (Berl.)\u003c/em\u003e \u003cstrong\u003e141\u003c/strong\u003e, 697\u0026ndash;708 (2021).\u003c/li\u003e\n\u003cli\u003eL\u0026ouml;vestam, S. \u003cem\u003eet al.\u003c/em\u003e Assembly of recombinant tau into filaments identical to those of Alzheimer\u0026rsquo;s disease and chronic traumatic encephalopathy. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e76494 (2022).\u003c/li\u003e\n\u003cli\u003eDelaglio, F. \u003cem\u003eet al.\u003c/em\u003e NMRPipe: a multidimensional spectral processing system based on UNIX pipes. \u003cem\u003eJ. Biomol. NMR\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 277\u0026ndash;293 (1995).\u003c/li\u003e\n\u003cli\u003eLee, W., Tonelli, M. \u0026amp; Markley, J. L. NMRFAM-SPARKY: enhanced software for biomolecular NMR spectroscopy. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 1325\u0026ndash;1327 (2015).\u003c/li\u003e\n\u003cli\u003eWilliamson, M. P. Using chemical shift perturbation to characterise ligand binding. \u003cem\u003eProg. Nucl. Magn. Reson. Spectrosc.\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, 1\u0026ndash;16 (2013).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":false,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4152095/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4152095/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Reducing fibrous aggregates of protein tau is a possible strategy for halting progression of Alzheimer's dis-ease (AD). Previously we found that in vitro the D-peptide D-TLKIVWC disassembles tau fibrils from AD brains (AD-tau) into benign segments with no energy source present beyond ambient thermal agitation. This disassembly by a short peptide was unexpected, given that AD-tau is sufficiently stable to withstand disas-sembly in boiling SDS detergent. To consider D peptide-mediated disassembly as a potential therapeutic for AD, it is essential to understand the mechanism and energy source of the disassembly action. We find as-sembly of D-peptides into amyloid-like fibrils is essential for tau fibril disassembly. Cryo-EM and atomic force microscopy reveal that these D-peptide fibrils have a right-handed twist and embrace tau fibrils which have a left-handed twist. In binding to the AD-tau fibril, the oppositely twisted D-peptide fibril produces a strain, which is relieved by the disassembly of both fibrils. This strain-relief mechanism appears to operate in other examples of amyloid fibril disassembly and provides a new direction for the development of first-in-class therapeutics for amyloid diseases.","manuscriptTitle":"How short peptides can disassemble ultra-stable tau fibrils extracted from Alzheimer’s disease brain by a strain-relief mechanism","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-08 19:06:17","doi":"10.21203/rs.3.rs-4152095/v1","editorialEvents":[],"status":"published","journal":{"display":false,"email":"[email protected]","identity":"nature","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nature","sideBox":"Learn more about [Nature](http://www.nature.com/nature/)","snPcode":"","submissionUrl":"","title":"Nature","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"019c536e-7671-4abe-9131-ad1e310c217b","owner":[],"postedDate":"May 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":29855986,"name":"Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy"},{"id":29855987,"name":"Biological sciences/Neuroscience/Diseases of the nervous system/Alzheimer's disease"}],"tags":[],"updatedAt":"2025-07-10T09:10:06+00:00","versionOfRecord":{"articleIdentity":"rs-4152095","link":"https://doi.org/10.1038/s41586-025-09244-z","journal":{"identity":"nature","isVorOnly":false,"title":"Nature"},"publishedOn":"2025-07-09 04:00:00","publishedOnDateReadable":"July 9th, 2025"},"versionCreatedAt":"2024-05-08 19:06:17","video":"","vorDoi":"10.1038/s41586-025-09244-z","vorDoiUrl":"https://doi.org/10.1038/s41586-025-09244-z","workflowStages":[]},"version":"v1","identity":"rs-4152095","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4152095","identity":"rs-4152095","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}