Cryo-EM structure of ex vivo Sup35 yeast prion and the factors defining prion phenotype

Cryo-EM structure of ex vivo Sup35 yeast prion and the factors defining prion phenotype | 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 Research Article Cryo-EM structure of ex vivo Sup35 yeast prion and the factors defining prion phenotype Dergalev Alexander A., Chesnokov Yuri M., Burtseva Anna D., Mitkevich Olga V., and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8764517/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Human prions and amyloids exhibit structural diversity that correlates with different pathology manifestations. Yeast prions represent a convenient model for studying basic properties of prions and their strain diversity. The yeast Sup35 prion has multiple structural variants divided into two groups, “strong” and “weak”, differing in the frequency of fibril fragmentation and resulting nonsense-suppressor phenotype. Yet, the molecular origin of these distinctions remains unclear. Here, we show that the difference in fragmentation between strong- and weak-type amyloid fibrils correlates with their stability, rather than with the efficiency of chaperone binding. The fully protease-resistant part (residues 2–32) of [ PSI+ ] fibril core is a key for the prion phenotype and maintenance, whereas partially resistant part (33–72) stabilizes the “strong” prion structure, and is less important for the weak variant. Using prion purified from yeast, we established the cryo-EM structure of the strong [ PSI+ ] variant, comprising residues 2–64, with 2.7 Å resolution. Using in silico modeling based on this atomic structure, we demonstrate that known anti-prion mutations in Sup35 N-domain decrease the thermodynamic stability of the structure. Comparison of ex vivo Sup35 prion structure with that of Sup35 amyloids formed in vitro shows that the latter differ significantly from the prion structure propagating in yeast cells. Structural Biology prions amyloids Sup35 cryogenic electron microscopy Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Amyloids are filamentous protein polymers, which contain highly ordered core with cross-ꞵ structure and may also include structured and unstructured parts. Amyloids are primarily known as pathological agents related to some incurable degenerative diseases of mammals 1 , 2 – but, as it is now known, they are widespread across all kingdoms of life, and often play an important role in the normal physiology of various organisms 3 – 5 . Amyloids, which can either spontaneously infect multicellular organisms, or stably propagate in microorganisms across many generations, are called prions. The [ PSI+ ] prion of yeast S. cerevisiae represents heritable amyloid of the eRF3 translation termination factor, also known as the Sup35 protein 6 – 8 . When Sup35 switches into the prion form, translation termination efficiency decreases. Thus, the [ PSI+ ] prion is an epigenetic element with a loss-of-function phenotype 9 . The Sup35 protein is composed of three domains. The N-terminal domain (N, residues 2-123) is mainly responsible for the prion properties and rich in glutamine and asparagine (QN). The middle domain (M, residues 124–253) is rich in glutamic acid and lysine and includes the major target for Ssa1/2 chaperones (residues 143–164) 10 . The C-terminal domain (C, residues 254–685) performs an essential function in the translation termination. Many amyloids exhibit structural polymorphism – i.e., the ability of the same amyloid-forming protein to acquire many different cross-ꞵ structures. Different amyloid polymorphs of α-synuclein, tau protein, Aꞵ peptide, PrP and other human proteins were shown to be associated with different types of respective neurodegenerative disorders 11 – 16 . Although many such pathology-related amyloid structures are now established 17 , 18 , the mechanisms relating these structures to pathologic phenotypes remain poorly understood due to the high complexity of multicellular organisms. In contrast, baker’s yeast is a simple unicellular organism, where the phenotypic manifestations of prions and their relation to prion structures can be observed and studied more easily 9 , 19 . The [ PSI+ ] prion phenotype is usually detected using nonsense mutations in adenine biosynthesis pathway genes, which cause accumulation of red intermediate of adenine biosynthesis. The Sup35 prion state reduces the efficiency of translation termination allowing partial readthrough of mutant nonsense codons, and reverting colony color from red to some shades of pink or white, depending on the extent of Sup35 inactivation (nonsense suppressor phenotype). The [ PSI+ ] prion also has a high number of structural variants differing by the level of nonsense suppression that are usually roughly divided into “strong” and “weak” types based on colony color phenotype 20 , 21 . This criterion is not sufficiently strict, but we have shown that strong and weak variants represent two distinct classes that can be reliably distinguished by their proteinase K (PK) digestion patterns and opposite reaction to overproduction of SUP35 and HSP104 genes 22 . The most systematic study of the [ PSI+ ] prion variation was able to distinguish 23 variants 23 . However, only five of them, including one strong and four of weak variants, were obtained through Sup35 overproduction, while others were generated by prion passage on various Sup35 mutants. It is widely assumed that many, if not all, phenotypic features distinguishing strong and weak variants result from the more frequent fragmentation of the strong Sup35 prions, compared to weak ones by Hsp104/Ssa1/Sis1 chaperone mechanism in yeast 24 – 26 . If so, the study of structure-to-phenotype relation in case of [ PSI+ ] prion variants could be reduced to a significant extent to seeking the mechanism relating the structures of Sup35 prions and the frequency of their fragmentation in yeast. Current understanding of the properties of Sup35 prions is largely based on studies of the amyloid fibrils spontaneously formed in vitro by the Sup35NM protein lacking the C domain. The fibrils formed at 4˚C produce mainly strong [ PSI+ ] prion upon transfection into yeast cells, while those formed at 37˚C produce mainly weak [ PSI+ ] 27,28 . Due to this, it is commonly assumed that the former fibrils (Sc4) are structurally equivalent to the strong prion variant, while the latter (Sc37) are equivalent to the weak one. Deuterium exchange and proteinase K (PK) digestion data indicate that the Sup35 amyloid cores include amino acid residues 2–46 in Sc4 and 2–72 in Sc37 29,30 . The Sc4 fibrils also showed significantly higher fragility than Sc37 28 . This allowed thinking that the higher fragility of strong prion variants defines their easier and more frequent fragmentation compared to weak ones. An alternative idea was that Hsp104 is sufficiently powerful to fragment any prion, but the critical point is the ability of the chaperone machinery to recognize prion as a target, and smaller prion core of strong prions leaves longer unfolded regions for better chaperone recognition 31 . However, PK mapping of prion cores of 26 strong and weak Sup35 prions isolated from yeast revealed that they all have a core structure comprising residues 2–72. However, while residues 2–32 are fully protected, the residues 33–72 are protected only partly and significantly better protected in strong variants 22 . This causes a paradox that by PK maps strong [ PSI+ ] are highly similar to Sc37 amyloid, rather than to Sc4, while weak [ PSI+ ] is similar to Sc4 22,30 . Current work, together with that of Tanaka et al. 32 resolves this paradox by showing the substantial structural difference of the Sup35 amyloids obtained in vitro and prion polymers propagating in yeast cells. Using the cryo-EM approach together with biochemical, computational, and genetic methods, we established the 3D-structure of the N-terminal core of strong-type Sup35 amyloid fibrils, analyzed the impact of different parts of Core 1 on the overall stability of these fibrils, and an effect of different mutations earlier reported to have anti-prion action. Based on these findings, we propose a model explaining how different Sup35 structures predetermine the different frequency of fibril fragmentation in vivo , and, eventually, the different severity of the [ PSI+ ] prion phenotype. Results 1. [ PSI+ ]-S7 and [ PSI+ ]-W2 prion fibrils differ in fragmentation due to their physical properties, rather than chaperone binding To study the structural basis of the difference of strong and weak [ PSI +] variants, we took [ PSI +]-S7 and [ PSI +]-W2 as the typical representatives of the “strong” and “weak” classes characterized in our previous work 22 . Consistent with earlier observations made by our and other groups on strong and weak [ PSI +] variants 21 , 33 , 34 , the cells bearing strong [ PSI +]-S7 variant had smaller size of Sup35 polymers, lower proportion of Sup35 monomers, and higher number of propagons (Fig. 1 A- 1 C, S1A) compared to the weak [ PSI +]-W2 variant. We compared the mechanical properties of Sup35 prions from [ PSI +]-S7 and [ PSI +]-W2 cells and their interaction with chaperones. The [ PSI +]-S7 prion was less resistant to heating and high-molar urea (Fig. 1 D- 1 E, S1C), and more fragile (Fig. 1 F). Assaying the amyloid/chaperone co-sedimentation, we observed that Sup35 fibrils of [ PSI +]-S7 and [ PSI +]-W2 variants bind similar amounts of the Hsp104 chaperone (Fig. 2 D), the key player in fragmentation and disaggregation of Sup35 polymers 24 , 26 . This agrees with the finding that strong- and weak-type Sup35 prion polymers bind equal amounts of the Ssa1/2 chaperones 35 . These results indicate that strong- and weak-type polymers represent equally attractive substrates for the chaperones performing amyloid fragmentation. Summing up, our data suggest that the fibrils stability is the primary feature that determines the frequency of their fragmentation, the balance between soluble and amyloid Sup35 protein and, eventually, the "strength" of the [ PSI +] prion phenotype. A. Phenotype of the strong and weak [ PSI +] variants in the ade1-14 yeast strain. B. Comparison of the Sup35 polymer size by the SDD-AGE and immunoblotting. C. Comparison of the proportion of soluble Sup35 in the cells carrying strong and weak [ PSI +] variants, determined as a proportion of Sup35 molecules capable of entering a 10% polyacrylamide gel without boiling. D-F . Comparison of resistance to heat (thermal stability assay, D ), urea ( E ) and mechanical shearing via ultrasonication ( F ) for [PSI+]-S7 and [ PSI+ ]-W2 prion fibrils. In D , the fractions of Sup35 prion fibrils, which resisted the indicated temperatures, were determined by SDD-AGE. In E , purified Sup35NM-GFP fibrils were incubated for the indicated time in 8 M urea, and the proportion of released monomeric Sup35NM-GFP was determined by SDS-PAGE, using a sample boiled without urea as a measure of the total Sup35NM-GFP amount in fibrils. In F , equal aliquots of Sup35NM-GFP fibrils were ultrasonicated for different durations, and the change in fibril size distribution was assessed by SDD-AGE. In all cases, electrophoresis was followed by immunoblotting and densitometry of blots. G. Hsp104 binding to strong- and weak-type Sup35 prion aggregates assessed via their co-sedimentation from cell lysates. The relative quantity of Hsp104 and aggregated Sup35 in pellets was assessed by Western blotting, using [ psi -] cells as a negative control. P-values were calculated using two-tailed Student’s t-test. 2. The role of Core1A and Core1B in the maintenance of Sup35 fibrils stability Since both strong and weak prion variants have the Core1 containing parts of similar length (residues 2–72 overall), the observed difference in physical properties of strong- and weak-type amyloid fibrils in vivo could not be explained by the different length of Core1. Paradoxically, the strong-type Sup35 fibrils, which are less stable than weak-type ones (Fig. 1 D- 1 F), and more fragmentable in vivo (Fig. 1 B), always have a more PK-resistant Core1B 22 than weak-type ones (which even can lack Core1B in rare cases). To explain this discrepancy, we hypothesize that in fibrils of strong prion variant Core1B plays an important role in stabilization of Core1A structure (which, itself, is inherently less stable), and appears to be more tightly associated with Core1A. In contrast, “weak” fibrils have a more stable fold of Core1A, and therefore Core1B is dispensable for the maintenance of their integrity (and could be more loosely attached to Core1A). To test this hypothesis, we, firstly, analyzed the effects on the [ PSI +]-S7 and [ PSI +]-W2 prions of several uncharged-to-charged amino acid substitutions in the Core1A and Core1B areas, previously described as prion-destabilizing ones. To this end, we substituted the wild type SUP35 allele by the mutant ones in [ PSI +]-S7 or [ PSI +]-W2 cells by the plasmid shuffle. The G58D substitution, located in the Core1B area (Fig. 2 A), was previously shown to cure only strong [ PSI +] variants 36 , 37 . Therefore, we anticipated that this mutation could destabilize either Core1B itself or its association with Core1A. Indeed, in our hands, SUP35 G58D allele, even as a sole source of Sup35 protein in cells, did not completely cure the [ PSI +]-S7 prion, but significantly altered its properties: weakened its nonsense-suppressor phenotype (Fig. 2 B, S2A), increased proportion of soluble Sup35 (Fig. 2 D), and decreased mitotic stability (Fig. 3 H). The PK resistance of the Core1B was noticeably reduced (Fig. 2 E). These features may seem to indicate a shift in the [ PSI +]-S7 G58D properties towards a weak prion variant. However, contrary to such interpretation, [ PSI +]-S7 G58D polymers became less thermoresistant and more fragile compared to [ PSI +]-S7 WT (Fig. 2 F- 2 G) - and, therefore, their manifestations in vivo likely originated from the increased (rather than decreased, as in the case of “true” weak [ PSI +]) fragmentation by chaperones; Fig. S2H), leading to a partial amyloid disaggregation, as it was initially proposed by Pei et al 38 . These data support the hypothesis that the Core 1B region is important for the in vivo stability of strong [ PSI +]. Reverse shuffle (Sup35 G58D → Sup35 WT ) led to restoration of the initial [ PSI +]-S7 phenotype (Fig. S2D-S2E), suggesting that Core 1B has regenerated its initial fold that is likely dictated by Core 1A. In contrast to the strong [ PSI +]-S7 prion, weak [ PSI +]-W2 variant was nearly unaffected by the Sup35 WT → Sup35 G58D substitution, as follows from its PK resistance profile, nonsense suppressor phenotype, and physical properties of fibrils (Fig. 2 B- 2 H, S2A). This suggests that Core1B is more loosely packed in the [ PSI +]-W2 fibrils, and, is more dispensable for structure stabilization and fibril’s physical properties. Earlier, we observed that a weak [ PSI +]-C3 isolate lacked protease resistance after residue Y45 39 . However, its thermal stability was even higher (Fig. S2F-S2G) than that of [ PSI +]-W2. This allows us to conclude that the region 46–72, a major part of the Core1B, is unlikely to have a significant impact on the overall stability of weak-type fibrils. Anti-prion substitutions S17R and Q24R 40 are located in the Core1A area (Fig. 2 A). The Sup35 WT → Sup35 S17R shuffle cured both [ PSI +]-S7 and [ PSI +]-W2 prions (Fig. 2 B). The Sup35 WT → Sup35 Q24R substitution also cured [ PSI +]-W2 variant, while [ PSI +]-S7 prion has acquired an unusual very weak nonsense-suppressor phenotype (dark-pink colony color, Fig. 2 B) with the absence of SDS-resistant polymers (Fig. 2 C). Since this phenotype did not emerge in the [ psi- ] cells Sup35 WT → Sup35 Q24R shuffle (Fig. S2B), and it was partially sensitive to Hsp104 inhibition (Fig. S2C), we suggest that it likely has prion nature, but the its prion structure is substantially altered and further study of its properties is beyond the scope of this article. Therefore, we conclude that Core1A is of ultimate importance for maintenance of both strong and weak prions variants, while Core1B fold is dispensable for weak variants and important for strong variant as a stabilizer of strong-type prion polymers. Based on these conclusions, we suggest that Core1A itself has less robust and stable fold in the strong-type fibrils (compared to weak-type ones), and therefore, requires stabilizing interactions with Core1B. A. Schematic representation of the Sup35 N-domain showing its Q/N-rich and oligopeptide repeats (R1-R4) regions, location of the fully PK-resistant Core1A and partly PK-resistant Core1B, and the studied anti-prion mutations. B. Nonsense-suppressor and Ade+ phenotypes of [ PSI+ ]-S7 and [ PSI+ ]-W2 cells with the SUP35 WT gene shuffled for either SUP35 G58D , SUP35 Q24R or SUP35 S17R mutant alleles. C. SDS-resistant Sup35 polymers detected by SDD-AGE in the lysates of [ PSI+ ]-S7 and [ PSI+ ]-W2 cells where SUP35 WT was shuffled for the indicated mutant variants. D. Proportions of soluble Sup35 and Sup35 G58D in [ PSI+ ]-S7 and [ PSI+ ]-W2 cells, estimated as by the Sup35 ability to enter a 10% polyacrylamide gel without boiling. E. Mapping of PK-resistant cores of prion fibrils isolated from [ PSI+ ]-S7 and [ PSI+ ]-W2 cells and composed of either Sup35NM WT -GFP or Sup35NM G58D -GFP proteins. F-G. Comparison of the resistance to heat (thermal stability assay, F ) and mechanical shearing by ultrasonication ( G ) for strong- and weak-type prion polymers consisting of either Sup35NM WT -GFP or Sup35NM G58D -GFP proteins. H. Comparison of prion mitotic stability for [ PSI+ ]-S7 and [ PSI+ ]-W2 cells with the SUP35 WT allele replaced for the SUP35 G58D or SUP35 Q24R alleles. P-values were calculated using two-tailed Student’s t-test 3. Cryo-EM analysis of [ PSI+ ]-S7 prion structure Next, we sought to resolve the atomic structures of ex vivo Sup35NM-GFP fibrils purified from either [ PSI+ ]-S7 or [ PSI+ ]-W2 cells. For both variants, most of the fibrils consisted of a single protofilament of ~ 6 nm width. For [ PSI+ ]-S7 fibrils, we observed alteration of nearly untwisted (~ 54%) and clearly twisted (~ 46%) segments on the same filaments (Fig. S4C). These segments displayed similar 2D class features apart from the degree of twist, indicating a common molecular fold. [ PSI+ ]-W2 fibrils had an almost untwisted morphology, which made them unsuitable for further structure reconstruction. Using the clearly twisted segments, we resolved the [ PSI+ ]-S7 fibril core structure at 2.7 Å resolution (Fig. 3 A- 3 H, S6A-S6D). The corresponding helical parameters were determined as a rise of 4.78 Å and a twist of − 1.39°, corresponding to a left-handed pitch of 123.5 nm. A single protomer exhibits a non-planar, undulating arrangement (with an axial span of ~ 9.1 Å along the fibril axis) and forms an “amyloid key”-like fold 17 in the S2-N64 region. This span of the ordered structure is generally consistent with the length of [ PSI+ ]-S7 Core1 as determined by PK mapping (Fig. 3 I). Residues 2–4 precede the structured region and are partially stabilized. Among them, residue 4 is sufficiently constrained for its position to be assigned unambiguously, whereas residues 2 and 3 remain partially flexible, contributing diffuse density without discernible side-chain features. The protomer molecule is stabilized by numerous “horizontal” interactions (Fig. 3 E- 3 F) between amino acid side chains, which stabilize the structure in the plane perpendicular to the fibril axis. An “inner” zipper, formed by a short β-arch spanning Y13–G25, stabilizes the central Q/N-rich subdomain of the core (Core1A). An “outer” zipper is generated by the antiparallel packing of N8–Q18 against Q47–Y63, thereby linking Core1A to the peripheral subdomain (Core1B) and integrating both into a single ordered fold. Because the protomer is non-planar, these interfaces are volumetric rather than flat: side chains interdigitate with a slight axial offset, producing three-dimensional packing that reduces solvent access and increases shape complementarity (Fig. 3 F). In both zippers, the antiparallel β-strands are staggered by approximately half of the helical rise, a feature typical of amyloid steric zippers that promotes volumetric packing. Quantitative analysis of interface geometry shows that the inner and outer steric zipper interfaces both display high shape complementarity (SC) 41 values of 0.81–0.82, in contrast to the lower value (0.76) displayed by the interfaces located outside of the zipper-forming regions (Fig. S6G). Notably, the SC values observed for the [ PSI+ ]-S7 steric zippers approach those reported for the microcrystals formed by the classical Sup35-derived peptide GNNQQNY (SC = 0.86), where the highly complementary polar side chains form completely dry steric zipper interfaces 42 . The order of the corresponding sidechains packing is also similar between the GNNQQNY microcrystal and the outer zipper of [ PSI+ ]-S7 amyloid (Fig. S6H). Along the fibril axis, stabilization is provided by both canonical in-register cross-β hydrogen-bonding network between backbone amides and carbonyls, as well as by some “vertical” interactions of the sidechains: twelve tyrosine residues form intermolecular 𝛑-columns, while numerous asparagine and glutamine residues form amide ladders (Fig. 3 G). A more detailed description of fibril-stabilizing interactions and some additional structural features is provided in the Supplementary Results section. Notably, the major PK-resistant peptides from [ PSI+ ]-S7 digestion start from the N-terminus (S2 residue) and end at the positions Y32, Y35, Q38, A42 and Y45 22 , which belong to the region between the “inner” and “outer” steric zippers (Fig. 2 H). As the core structure predominantly consists of polar residues (Fig. 3 D), its surface is largely hydrophilic and uncharged (Fig. S6E-S6F). A. Cross-sectional view of the 3D reconstruction of [ PSI +]-S7 amyloid core structure. B-C. Rendered 3D model of the ordered part of [ PSI +]-S7 amyloid fibril ( B ), and the atomic model of a single protomer superimposed on the sharpened cryo-EM density map ( C ). D. The topology of the [ PSI +]-S7 amyloid fold, showing the location of β-strands. The sidechains are color-coded based on their physiochemical properties. E. Atomic model of the [ PSI +]-S7 amyloid core showing the “inner” and “outer” steric zipper elements. F. Close-up of the parts of “inner” and “outer” steric zippers showing their “staggered” organization along the Z-axis of the fibril. G. Close-up of several 𝛑-columns formed by tyrosine residues (green arrows) and amide ladders formed by glutamine residues (violet arrows) stabilizing the core structure along the axis of [ PSI +]-S7 prion fibrils. H-I. Comparison of the PK-protected region and cryo-EM-solved ordered structure of Core1. Dotted lines indicate the main stop sites for proteinase K digestion of Core1B. The “inner” and “outer” steric zipper zones are highlighted in magenta and blue, respectively. The major PK-resistant peptides observed in the [ PSI+ ]-S7 preparation are shown below. 4. Thermodynamic profiling of [ PSI+ ]-S7 atomic structure and its interaction with the anti-prion point mutations in Sup35 N-domain. To make a structure-based assessment of the [ PSI+ ]-S7 fibrils stability, we estimated their free energy of unfolding ( ΔG Unf ) using the FoldX algorithm 43 . For comparison, we also calculated the solvation energy ( ΔG Solv ), which has been used to estimate the amyloid structures stability in many published works 44 – 48 . To get a broader picture, we also calculated ΔG Unf and ΔG Solv for a set of published amyloid structures belonging to different partially overlapping biological and physicochemical groups. We found that [ PSI+ ]-S7 structure has ΔG Unf comparable to that of some structures of known functional amyloids, and higher (i.e. less favorable) than that of pathology-related amyloids (Fig. 4 A). In contrast, its ΔG Solv is very high (i.e. non-favorable) even in comparison with the majority of functional amyloids. Notably, the only known amyloid structure with even higher ΔG Solv compared with Sup35 fibrils is the Orb2 functional amyloid of D. melanogaster 49 . Given that both of these structures are dominated by stabilizing steric zipper motifs (as in the case with many other highly stable pathological amyloid structures), we suggest that the ΔG Solv parameter is much less accurate for assessing amyloid structure stability than the ΔG Unf parameter, and seems to greatly underestimate the true stability of amyloid structures consisting mostly of polar residues (Fig. S7). We further used the resolved [ PSI+ ]-S7 amyloid core structure to provide structural rationale for the effect of known anti-prion mutations in Sup35 N-terminal domain. To this end, we used a set of 15 such mutations that mostly cause uncharged-to-charged residue changes 40 , 50 , 51 . A majority of these substitutions are located in the areas of either “inner” or “outer” steric zippers (Fig. 4 B), suggesting that the zipper regions are most important for the structure stability. Using FoldX, we estimated the fibril stability change upon these mutations, and found all substitutions to be significantly destabilizing (i.e., increasing ΔG Unf ; Fig. 4 C). Remarkably, by testing another 10 manually selected uncharged-to-charged residue substitutions located in non-zipper-forming regions, we found their effect on [ PSI+ ]-S7 structure stabilization energy to be negligible. Thus, [ PSI+ ]-S7 structural analysis indicates which structural motifs are most sensitive to substitutions. One could also suggest that introducing charged residues into the tightly locked zippers would loosen their packing, thereby decreasing fibril stability (Fig. 6 C). A. Gibbs free energy of unfolding and solvation energy of Sup35/[ PSI+ ]-S7 prion structure (red arrow) in comparison with that of several representatives of different amyloid classes. B. Amyloid core structure with known anti-prion mutations 40 , 50 , 51 . Most of them represent uncharged-to-charged residues substitutions in the zipper-forming regions. C. Predicted change in the free energy of unfolding (ΔΔG) calculated by FoldX software for the anti-prion Sup35 mutations in comparison with the wild type structure. In contrast to known anti-prion mutations, manually selected substitutions (uncharged-to-charged residues) located outside of the zipper-forming regions have only a minor impact on prion stability. P-value was calculated using two-tailed Student’s t-test. 5. Comparison of ex vivo Sup35 prion structure with in vitro formed Sup35 amyloids. The work of Tanaka et al. published recently 32 , describes the cryo-EM structures of Sc4 and Sc37 amyloids that are presumed to be structural equivalents of the strong and weak Sup35 prions, respectively. Comparison of the structures of Sc4 amyloid with that of the strong [ PSI+ ]-S7 prion reveals both a significant similarity and a substantial difference (Fig. 5A-5B). The similarity relates to a region 2–26, which we consider to be the most significant and defining part of the [ PSI+ ]-S7 structure. The highest similarity is observed in the area of Y13–G25 β-arch (Fig. 5B, S6G). A ꞵ-strand N27-Q30 is also found in both structures, but it is oriented differently. In contrast to [ PSI+ ]-S7, in Sc4 structure the major part of the N36-N64 stretch (roughly corresponding to the Core1B area) is fully or partially disordered. The exception is the Q47-S53 stretch, which forms short ꞵ-strand attached to the top of Y13-G25 hairpin by the zipper-like interactions. However, the registry of these interactions in [ PSI+ ]-S7 and Sc4 structures is different. Surprisingly, the structure of Sc37 amyloid also bears noticeable similarity to that of [ PSI+ ]-S7 prion: both structures contain long zipper-like element between the antiparallel stretches Q6-Q23 and Y45-N63 (Fig. 5A, 5C, S6G), and similar overall span of the Core1 (residues 5–65 and 2–64, respectively). The thermodynamic parameters ( ΔG Unf and ΔG Solv ) are also similar between [ PSI+ ]-S7 prion and Sc37 amyloid structures, as well as between Sc4 and the Core1A part of [ PSI+ ]-S7 structure (Fig. 5D). Tanaka et al. also reported that about a half of the Sc37 amyloid segments also contained the additional core (K102-Q132); this subtype of Sc37 structure has ΔG Unf ~25% lower compared with that of [ PSI+ ]-S7 structure. Figure 6 . Comparison of ex vivo Sup35 prion structure with in vitro formed Sup35 amyloids A. Structure of [ PSI+ ]-S7 ex vivo prion fibrils in comparison with the Sc4 (PDB ID: 9XBN) and Sc37 (PDB ID: 9XBO) amyloids formed in vitro 27 , 28 . Steric zipper elements are highlighted in grey. B-C. Structural alignment of [ PSI+ ]-S7 with Sc4 ( B ) and Sc37 ( C ) structures, with the close-up of some regions fully or partially similar between [ PSI+ ]-S7 with Sc4 structures. D. Comparison of the free energy of unfolding ( ΔG Unf ) and solvation energy ( ΔG Solv ) between [ PSI+ ]-S7, Sc4 and Sc37 structures. Discussion The mechanism of prion fragmentation Previously, two hypotheses were proposed explaining what defines the difference in the frequency of fragmentation between the weak and strong Sup35 prions. One hypothesis proposed that the more frequent fragmentation of the strong prion is due to its lower mechanical strength. The other assumed that since Hsp104 is a powerful molecular machine with 12 ATPase units, the limiting factor is not its ability to unfold a prion, but its ability to find its target, which occurs with the help of Hsp70 (Ssa1/2) and Hsp40 (Sis1). In support of the latter model, it was observed that progressive deletions of the oligopeptide repeats in the Sup35 N domain, which were assumed to be chaperone targets, reduced prion fragmentation, thus increasing prion size and eventually caused prion loss 52 , 53 . However, the set of data, presented in this and several other works 31 , 35 , 54 , supports the mechanical strength being the key factor. This also agrees with the recent finding that the Sup35 interaction with chaperones mainly occurs through the Ssa1/2 binding site located beyond the prion-forming region, at Sup35 residues 143-164 10 . While one can note that the mechanical strength model was generally more popular than its alternative, it was based predominantly on the assumption of the equivalence of synthetic Sup35 amyloids and in vivo Sup35 prions, which proves to be incorrect according to this and some previous studies. Cryo-EM structure of Sup35 prion Many mammalian amyloid fibrils are composed of two or more protofilaments. In contrast, the Sup35 polymers extracted from both [ PSI +]-S7 and [ PSI +]-W2 cells consisted of a single filament. This rules out the scenario where the higher stability of weak [PSI+] filaments is related to their double- or triple-protofilament arrangement. In vitro , Sup35 usually forms single filaments, though twin filaments were also observed 32 . However, it appears likely that only single-filament fibrils could succeed as viable prions in yeasts. Propagation of yeast prions relies on the ability of Hsp104 to fragment prion fibers 55 . While in single-filament fibers this requires the extraction of just one protomer, in multi-protofilament fibers several protomers opposing each other should be extracted. Here we found that [ PSI+ ]-S7 prion has the parallel-in-register architecture. This confirms the earlier conclusion of Wickner and coauthors 56 – 58 , and disproves the alternative solenoid-type model 59 , 60 . Of note, the solenoid model was formulated basing on study of in vitro generated Sup35 fibrils 59 , for which the parallel in-register architecture was also shown recently 32 . Importantly, PK digestion patterns of strong and weak prions are similar within these groups 22 , suggesting that many (if not all) in vivo [ PSI+ ] variants are likely to share similar strong or weak fibril architecture. In this work we showed that the Core1A (residues 2–32) is the key element of the strong and weak prion structures. While some known anti-prion substitutions are located in the Core1B region 50 , genetic screen performed by DePace et al yielded the anti-prion substitutions located mostly between residues 8 and 24 40 , which corresponds well with the location of inner and outer steric zippers (Fig. 3 E, 4 B). Our data (Fig. S2D) and a previous report 61 suggest that the Core1A region affects the folding of the rest of the structure. While the dominance of the Core1A appears to be largely related to its physical strength, our recent study shows that the terminal location itself is a significant factor helping any sequence to influence folding of the remaining part of prion structure. The most telling observation was that a random 23 residue sequence added to the Sup35 N-terminus altered the fold and overall span of the Sup35 prion structures in their original location 62 . Relation between the in vitro formed Sup35 amyloids and the prion propagating in vivo . Earlier works revealed a significant difference of the PK maps of the strong and weak Sup35 prions with their presumed counterparts generated in vitro 22 , 30 . This work, together with Tanaka et al 32 resolves this discrepancy by establishing cryo-EM structures and showing that the PK maps correspond well to these structures, while the structures differ significantly. The Sc4 and [ PSI+ ]-S7 structures are very similar in their most important fully PK-resistant part, or Core1A, but have little in common for the rest of the structures. Therefore, when yeast is transfected with Sc4 amyloid, its region 32–64 has to rearrange, acquiring the continuous cross-β fold (Fig. 6 A). Alternatively, a conformer similar to the [ PSI+ ]-S7 prion may be present in Sc4 preparations as a minor fraction, becoming the most abundant in yeast (Fig. 6 B). A recent study using in vivo NMR indicates that both of these processes occur with in vitro -formed α-synuclein amyloid: upon transfection into cells, the proportion of major and minor isoforms reversed, and the disordered flanking regions of the fibrils were remodelled 63 . Our observations with the Sup35 G58D anti-prion suggest why the [ PSI+ ]-S7 fold may have selective preference in yeast cells. This substitution destabilizes the steric zipper which attaches Core1B to the Core1A, making the whole prion structure less stable (Fig. 2 E- 2 G, 4 B- 4 C), which should increase prion disaggregation and promote prion loss 38 . Apparently, the Sc4 structure, which almost lacks this steric zipper, should be even less stable than the strong prion with G58D substitution and thus unable to propagate in yeast. Although the cryo-EM structure of the weak [ PSI+ ] variant has not been solved, some important notes about the relation of the Sc37 structure and the natural weak [ PSI+ ] prions can be made. Comparison of the Sc37 and [ PSI+ ]-S7 structures explains why their PK digestion patterns are so similar. Both structures include similar, though not identical steric zippers formed by stretches N8-Y16 and Q50-Y63, which ensures high resistance to PK of the region Y45-Q72 in both cases. However, in all studied weak [ PSI+ ] prions, and [ PSI+ ]-W2 in particular, this region shows low PK resistance 22 . Summing up, in contrast to the Sc4 case, it is difficult to see an easy way for the Sc37 structure to rearrange into a weak prion structure. It was shown that Sc37 transfection into yeast produces a weak [ PSI+ ] variant similar to the VK variant 23 , which exhibits a PK digestion pattern typical for all weak variants 22 , but different from that of Sc37 30 . Thus, it appears likely that Sc37 preparation contains a rare polymorph, structurally distinct from the published Sc37 major structure, that can be preferentially selected in cells and give rise to a weak [ PSI+ ] prion. Relation between the length of Sup35 amyloid core structures and PK-resistant peptides. Mapping of amyloid cores through partial proteinase K digestion is an affordable supplement and in some senses an alternative to the full-atomic structure reconstruction. Now, PK maps are available for several ex vivo yeast prions 62 , 64 , but their full-atomic structures are missing. Thus, it is of interest to understand how PK maps relate to actual structures. This work and those of Tanaka group 30 , 32 provide a rare opportunity to learn this. The ordered core of [ PSI+ ]-S7 fibrils defined by the cryo-EM (residues 2–64) is smaller than the PK-resistant region (residues 2–72). Similar differences were observed for the Sc37 amyloid (5–65 versus 2–72) 30 , 32 . Sc4 has an ordered core comprising residues 4–35 and 47–52, while the PK-resistant region includes residues 2–46; cryo-EM structure of S17R4 amyloid includes residues 88–129 versus PK-resistant core at residues 81–148; S17R37 cryo-EM structure includes residues 69–132 and PK-resistant core at residues 62–144. Thus, PK-protected regions are about 7–12 residues longer at each end excluding the N-terminus, which has no such length of unstructured sequence. The existence of such a gap could be explained by the fact that the PK active site is located within a pit of the PK molecule 65 , and to be digested at its end, an unfolded region of certain length needs to enter this site and to reach the catalytic center. Still, it is unclear how PK can cut within the structured region at residues 32 to 64, and this mechanism requires further investigation. A-B. Two possible explanations of relation between the structures of in vitro formed Sc4 amyloid and in vivo [ PSI+ ]-strong prion variant. First one ( A ) implies structural rearrangement of Sc4 amyloid by the extension of existing cross-β fold on some unfolded parts of Sup35 molecule (the N36-Y46 and G54-N64 stretches), and the second one ( B ) implies positive selection of rare, but the most fitted Sc4 polymorphs inside the yeast cells. C. Proposed mechanism of the anti-prion action of the G58D and Q24R substitutions with respect to the strong [ PSI+ ] variant structure resolved in this work. Materials and methods Yeast strains and cultivation conditions The work used derivatives of the 74-D694 yeast strain ( MATa, ade1-14, trp1-289, his3Δ-200, ura3-52, leu2-3,112 ) carrying either [ PSI +]-S7 or [ PSI +]-W2 prion. To prevent the de novo formation of any alternative amyloid conformers in the experiments with Sup35NM-GFP overproduction, the RNQ1 gene in these strains was disrupted by HIS3 insertion. For the plasmid shuffle experiments, derivatives of [ PSI +]-S7 and [ PSI +]-W2 strains were created, where the chromosomal SUP35 gene was disrupted by TRP1 insertion and the SUP35 gene introduced on the centromeric pRS316-SUP35 vector. The standard yeast media were used. Synthetic complete media (SCM) contained 6.7 g/L yeast nitrogen base, 20 g/L glucose or 24 g/L galactose, and required amino acids. For colony color development, SCM contained reduced amount of adenine (7 mg/L, or 1/3 of standard). For overproduction of proteins of interest under control of the GAL1 promoter, we used SCM-Gal medium with 24 g/L galactose instead of glucose and 100 mg/L of adenine. Rich YPD medium contained 10 g/L yeast extract, 20 g/L peptone, 20 g/L glucose and 20 g/L agar when required. For better development of the colony color phenotype, solid YPD-red medium was used (6 g/L yeast extract (Oxoid), 20 g/L peptone, 20 g/L glucose and 20 g/L agar). Cells were grown at 30°C. Yeast genetics assays Yeast transformation with DNA was performed using the standard LiAc-PEG method 66 . Yeast transfection with prion fibrils (“spheroplast transformation”) was performed as described by Tanaka et al 27 , using an “empty” pRS316 vector as a selection marker. To switch the Sup35 WT allele to Sup35 MUT , we used the plasmid shuffle technology. Cells where the genomic SUP35 was deleted, and the episomal Sup35 WT was encoded on the centromeric pRS316 ( URA3 ) plasmid, were transformed with pRS315 ( LEU2 ) plasmid encoding the Sup35 MUT allele. These SUP35 genes included the native promoter. Transformants were grown on selective medium for 3 days, and then streaked on the complete synthetic medium containing 1 mg/ml of fluoroorotic acid (FOA) to ensure loss of the URA3 plasmid. After 3-day growth, the cells from fully-grown colonies were pooled, streaked to single cells on the YPD-red medium, grown for 3 days, and the loss of plasmid encoding the Sup35 WT allele was confirmed by inability to grow on the synthetic medium lacking uracil. To make yeast spots, cells were grown overnight in liquid YPD, then cell suspensions were diluted to OD600 = 1, and 8 µl of suspensions were spotted to the YPD-red plates. For better development of colony color, spots were grown at 30°C for 3 days and then incubated at 4°C for 1 day prior imaging. The propagon number in [ PSI +] cells was evaluated as described by Cox et al 34 . Briefly, cells were streaked on YPD plates, grown at 30°C for 3 days, freshly-grown colonies were streaked on YPD plates with 4 mM GuHCl and grown for 1.5-2 days. Small colonies were collected and streaked on the complete synthetic medium containing 0.4 mg/ml adenine (2% of standard concentration). At this adenine concentration, only the cells that inherited [ PSI +] prion can form fully-grown colonies. Plates were incubated at 30°C for 5 days, and the number of fully-grown colonies corresponding to each colony from the YPD-GuHCl plate was calculated. It is assumed that this value corresponds to the number of propagons (the elementary units of prion inheritance) in a single yeast cell carrying [ PSI +] prion. To estimate the level of prion stability upon the mitotic cell growth, several [ PSI +] colonies were pooled, streaked to single cells on YPD plate and grown at 30°C for 3 days. Then 5 individual colonies (as individual biological replicates) were separately streaked to single cells on YPD-red plated, grown at 30°C for 3 days, then incubated 1 day at 4°C, and the proportion of red (i.e. [ psi -]) colonies were determined for each replicate. As the single cell takes ~ 25 generations to form a fully-grown colony, the obtained values correspond to the percentage of prion loss after 25 generations of mitotic growth. To demonstrate the increased level of the weak [ PSI +]-W2 prion loss upon the short-term heat shock, cells carrying [ PSI +]-S7 and [ PSI +]-W2 prions were grown in liquid YPD medium to the log phase (OD600 ~ 1.5), then incubated at 45°C for 30 min, streaked to the single cells on YPD-red plates and grown at 30°C for 3 days. The proportions of red and red/pink sectored colonies were calculated. Purification of Sup35NM-GFP prion fibrils from yeast for cryo-EM analysis. Method consideration and creating the protein construct In [ PSI +] cells with the native production level of endogenous Sup35 protein, prions exist as a population of short fibrils 33 , which are not suitable for the cryo-EM structure reconstruction. To overcome this issue and facilitate prion purification we overproduced the GFP-tagged construct containing prionogenic part of Sup35 protein (residues 1-239) using the multicopy plasmid carrying the SUP35NM-GFP gene under the control of GAL1 promoter. Upon this procedure, this construct adopts the pre-existing prion fold 22 and form long filaments sequestered into the large higher-order aggregate 67 . In this and previous work 68 we observed that Sup35NM-GFP fibrils per se are poorly suitable for cryo-EM reconstruction due to the structural noise imposed by GFP globules and the disordered Sup35 M domain (a. a. 125–239). These parts were removed by trypsin. However, the obtained fibrils were prone to precipitation, so that only a small proportion of them were separate on cryo-EM grids and suitable for structure reconstruction. To cause repulsion of fibrils, we introduced an acidic cluster DDDNEDSEEDDEDGGP R GSR between G96 and A155 residues, resulting in the Sup35DE-GFP protein (Figure S4A). Using MALDI, we confirmed that trypsin cleaves this protein at arginine residue (in bold), but does not cleave at the only upstream R28 located inside the prion core. Purification of Sup35DE-GFP fibrils Yeast cells carrying either [ PSI+ ]-S7 or [ PSI+ ]-W2 prions were transformed with two plasmids: the pYES2-SUP35DE-GFP plasmid for overproduction of prion protein, and the rescue plasmid pRS315-SUP35C which reduced the cellular toxicity associated with Sup35 overproduction in the [ PSI +] cells, thereby increasing the level of Sup35DE-GFP production. Cells were grown to the stationary phase in selective SCM-glucose medium, then a two-fold volume of selective SCM-Gal medium was added, cells were incubated for 15 hours and harvested. Cell pellets (2*4 ml in 50 ml tubes) were mixed with an equal volume of glass beads and a half volume of lysis buffer (TBS (Tris Buffered Saline) buffer, 5 mM PMSF, 1 mM DTT). Cells were broken by vortexing at maximum speed for 10 minutes at 4ºC. Cell lysates were transferred to 15-ml tubes and spun down at 2500g for 15 min at 4ºC. The supernatant lacked GFP and was discarded, and the upper layer of pellet containing GFP was resuspended in 4 ml of fresh TBS buffer supplemented with 100 µg/ml RNaseA, 50 µg/ml DNase I, placed on top of sucrose gradient #1 (TBS based, 0.5 ml of 60% sucrose and 3 ml of 30% sucrose in 15 ml Falcon tube), and spun for 15 min at 2500g, 4ºC. The GFP-containing fraction of gradient was collected and sucrose was diluted two-fold to reduce sucrose concentration. To disassemble higher-order clumps of Sup35DE-GFP prion polymers into individual fibrils, SDS was added to the final concentration of 4%. Samples were incubated for 20 min at 10ºC to avoid both protein degradation and SDS precipitation. The dissolution of visible Sup35DE-GFP clumps into individual fibrils was monitored with the fluorescent microscope Axioskop 40 (Zeiss, Germany). Then, samples were centrifuged for 1 min at 21300 g to remove the remaining debris, and GFP-containing supernatant was placed on top of sucrose gradient #2 (based on TBS buffer with 0.3% Sarcosyl: 100 µl of 70% sucrose + 200 µl of 60% sucrose + 700 µl of 30% sucrose in 3.5 ml tube) and spun down in Beckman Coulter Optima Max ultracentrifuge SW50 rotor at 260 000 g for 4 h at 10ºC. GFP-tagged prion fibrils were trapped in the 60% sucrose fraction. The GFP-positive fraction of the gradient was collected, frozen in liquid nitrogen and stored at -70ºC. To remove GFP and “fuzzy coat” part of Sup35DE-GFP fibrils, the samples were diluted two-fold with TBS buffer and treated with 0.1 mg/ml trypsin (Sigma) for 60 min at 37°C. To terminate reaction, 1 mM PMSF was added after 1 h incubation. To remove the unwanted digestion products, the preparations were centrifuged at 260000 g for 4 h at 4°C using sucrose gradient #2. Pure fibrils were found in the 60% sucrose fraction. This fraction was dialyzed against 1 L of 0.1X TBS buffer with 0.1% Sarcosyl. The samples were then concentrated ~ 10-fold by evaporation of water excess for 1 h at room temperature. The scheme of fibril isolation, trypsin treatment and purification is shown in Figures S4A and S4B. The purity of the final sample, as determined by the SDS-PAGE, was ~ 95% (Fig. S3C). The Sup35DE fibrils showed reasonably small aggregation on the cryo-EM grids, suitable for cryo-EM data collection. To ensure that prion strain properties were not changed upon Sup35DE-GFP overproduction and protein purification procedure, we transfected the 74-D694 [ psi -] cells with either [ PSI +]-S7 or [ PSI +]-W2 fibrils, extracted and treated as described above. In both cases, initial prion phenotype (either strong or weak nonsense suppression) was reproduced (Fig. S4B). Cryo-electron microscopy Sample preparation and data acquisition . Quantifoil R1.2/1.3 Cu grids were glow-discharged for 20 s at 20 mA (0.26 mbar) using a PELCO easiGlow system. 3 µl of the fibril suspension (0.38 mg/ml) were applied to the grids, blotted for 3 s at 100% humidity and 4°C, and plunge-frozen in liquid ethane using a Vitrobot Mark IV (Thermo Fisher Scientific, USA). The grid was then stored in liquid nitrogen until use. Cryo-EM data were collected on a Titan Krios 60–300 transmission electron microscope (Thermo Fisher Scientific, USA) equipped with a field emission electron gun X-FEG (Thermo Fisher Scientific, USA), spherical-aberration corrector (CEOS GmbH, Germany), a post-column BioQuantum energy filter (Gatan, USA) and a K3 direct electron detector (Gatan, USA) in counting mode using SerialEM 4.055 and 9-hole (2 exposures per hole) image-shift data acquisition strategy at the National Research Centre ‘‘Kurchatov Institute’’. The microscope was operated at 300 kV with a nominal magnification of 81000x, corresponding to a pixel size of 0.863 Å at the specimen level, and an electron energy selecting slit of 20 eV. A total dose of 52 e⁻/Ų within a 3.5 s exposure time was fractionated into 70 frames, resulting in an electron dose of 0.74 e⁻/Ų per frame. A total of 9,015 movies were collected in a nominal defocus range from − 0.7 to -2.0 µm with a step of 0.1 µm. Detailed parameters of data acquisition are listed in Table S1. Image processing. Movies were motion-corrected and CTF parameters estimated in Warp. Micrograph power spectra showed a characteristic ~ 4.8 Å meridional reflection, indicative of well-ordered cross-β structure. Filaments were automatically picked with crYOLO using an inter-box step of 24 Å (five helical units per step). Helical segments were extracted with box sizes of 1024 and 512 pixels (binned to 256 and 128 for the initial steps) to support accurate crossover measurements and efficient 2D classification, respectively. 1024-pixel and 512-pixel particles were used for initial 2D classification, helical pitch estimation, and ab initio model generation. The 2D classes from 1024-pixel boxes reported a crossover distance of ~ 70 nm. After initial 2D classification with 512-pixel boxes, false positives and broken filaments were removed and the dataset was partitioned into twisted (585,256 segments) and low-twist (674,012 segments) subsets. For homogeneous processing, only the twisted subset was retained. An initial 3D model of the twisted fibrils was then generated using relion_helix_inimodel2d , assuming a left-handed helix, and used as input for subsequent refinement. Starting from 585,256 segments extracted with a 384-pixel box, an initial helical refinement in CryoSPARC yielded a reconstruction at 3.6 Å, which improved to 3.23 Å after local refinement with a soft mask around the fibril core. This was followed by two consecutive rounds of 3D classification without alignment in CryoSPARC, each combined with cycles of CTF refinement, helical refinement, and local refinement. In the first round, a well-defined subset comprising 163,538 segments produced a 2.81 Å reconstruction (helical parameters: rise 4.78 Å, twist − 1.278°). A second round of classification of this subset was performed to further purge residual heterogeneity, yielding 50,658 high-quality segments. Refinement of this final subset in CryoSPARC resulted in a global resolution of 2.69 Å, as determined by the gold-standard FSC 0.143 criterion. The best-defined map corresponded to fibrils exhibiting a rise of 4.782 Å and a twist of − 1.388°. The map was sharpened with a negative B-factor of − 58 Ų. Model Building and Refinement. A single protomer was built de novo into the sharpened map in Coot, with unambiguous sequence assignment from residue 4 to residue 64. Density preceding residue 4 was poorly resolved: residues 2–3 were therefore modeled according to sequence context (initiator Met1 absent; Ser2 Nα-acetylated; Asp3 present), while acknowledging their flexibility in the map. The built protomer was stacked into a five-layer segment by rigid-body placement in UCSF Chimera, and subsequently subjected to PHENIX real-space refinement with NCS constraints between the five chains and individual ADPs. Model-to-map and stereochemical validation statistics are summarized in Table S1. Shape complementarity (SC) values were calculated using the sc utility from the CCP4 suite for selected pairs of residue segments forming lateral β-sheet interfaces, with molecular surfaces constructed from five neighboring fibril layers to reduce edge effects. PK resistance analysis of Sup35NM-GFP prion fibrils Yeast cells carrying either [ PSI+ ]-S7 or [ PSI+ ]-W2 prions were transformed with pYES2-SUP35NM-GFP plasmid (where Sup35NM sequence was either WT or containing G58D substitution) and grown to the stationary phase in the selective SCM-glucose medium. Then a two-fold volume of selective SCM-Gal medium was added, cells were incubated for 15 hours and harvested. To map the PK-resistant regions of Sup35NM-GFP fibrils, we extracted and purified the Sarcosyl-resistant fractions from large pellets of yeast cells (~ 3–5 grams per preparation), as described in the original work 22 . Briefly, yeast cells were disrupted with glass beads, fraction of Sup35NM-GFP aggregates was isolated, mixed with Sarkosyl (final concentration = 5%), sonicated, loaded onto a 20–65% sucrose gradient, and ultracentrifuged at 260000g for 4 hours at 4ºC. The GFP-containing fraction was collected, and the resulting Sup35NM-GFP sample (diluted to a final concentration of 200 µg/ml) was digested with PK (25 µg/ml) for 1 hour at RT. The PK-resistant part of the fibrils was then precipitated with 40% acetone, resuspended in water, boiled, and analyzed by MALDI-TOF/TOF mass spectrometer UltrafleXtreme (Bruker, Germany). PK has no sequence specificity except it does not cut before and after proline 22 , 62 . Peptides were identified by tandem mass spectrometry (MS-MS) and/or as groups of related peaks. As we have previously shown 22 , the N-terminal methionine of Sup35PD-GFP was completely removed and replaced by an acetyl group in the yeast cells, and thus all N-terminal peptides started from amino acid residue № 2. Raw MALDI data were analyzed using Bruker flexAnalysis 3.3 software, and PrK resistance was calculated and plotted as a function (Rn) of the normalized PrK resistance index for a given sequence position (n), given by: Rn = ∑ S n / ∑ S max where the ∑ S n is a sum of the areas of all MS peaks containing this amino acid position (n), and ∑ S max – the maximum peak area sum value. SDD-AGE analysis of amyloid polymers Yeast cells were grown in 20 ml of the YPD media to the mid-log phase, collected to 2.4 ml Eppendorf tubes and lysed by vigorous vortexing with glass beads and 100 µl of TBS buffer supplemented with 5 mM PMSF, 1 mM DTT and Complete protease inhibitor cocktail (Roche) at 4°C for 10 min. Cell lysates were mixed with 4X SDD-AGE sample buffer (1X is 0.5X Tris acetate/EDTA, 2% SDS, 5% glycerol, and 0.05% Bromophenol blue) and subjected to the agarose gel electrophoresis, as described by Kryndushkin et al 33 . Protein samples resolved in agarose gel were then vacuum-transferred to a nitrocellulose membrane (Porablot, MACHEREY-NAGEL), and subjected to a standard Western-Blotting procedure using anti-Sup35NM rabbit polyclonal primary antibodies. Thermal stability assay The procedure was performed as standard SDD-AGE, but with some modifications 31 . Cells were grown and lysed as described above. To avoid possible effect of cell lysate components on the thermal stability, the Sup35 aggregates were purified by centrifugation at 20000 rpm for 30 min at 4°C. The supernatant was discarded, and the pellet was dissolved at room temperature in 300 µl of SDD-AGE sample buffer. Remaining cell debris was spun down at 4000 rpm for 1 min, and 50 µl sample aliquots were transferred to PCR tubes and incubated at either 25, 50, 60, 70, 85, or 99°C for 8 minutes, then cooled to 10°C and loaded onto a 1.8% agarose gel (gel strength = 1000 g/sm 2 ) with 0.1% SDS and analyzed by SDD-AGE followed by Western blotting with anti-Sup35NM primary polyclonal antibodies. The integrated intensity of Sup35NM-GFP smears, as a measure of the amount of Sup35NM-GFP amyloids that withstood heating, was then quantified using Fiji software. For each amyloid preparation, the experiment was performed in 3–4 replicates, and mean ± SEM values were calculated. Mechanical fragmentation of Sup35NM-GFP fibrils The fibrils of Sup35NM-GFP (with either WT sequence of Sup35NM, or containing G58D substitution) were extracted from the yeast cells using the protocol similar to that used for fibril purification for the cryo-EM study: the big clumps of long fibrils were isolated from the yeast lysates, dissolved into single fibrils with 4% SDS, and the fraction of purified fibrils was then obtained by ultracentrifugation in sucrose gradient. The obtained fractions were dialyzed overnight against the TBS buffer, obtained samples were diluted 20-fold and divided into 0.5 ml aliquotes. One aliquot was leaved unsonicated, while others were sonicated (2/2 second on/off cycle, 50% amplitude) using VibraCell sonicator (Sonics & Material Inc, USA) for indicated times at + 4°C. All samples were then mixed with SDD-AGE sample buffer and subjected to 1.8% agarose gel electrophoresis, followed by Western blotting and immunostaining with anti-Sup35NM primary antibodies. The distributions of Sup35NM-GFP signal on the gel, which reflect the Sup35NM-GFP fibril mobility and the shift in their size upon sonication, were quantified using densitometry analysis with FIJI software. Urea denaturation assay Sup35NM-GFP fibrils, templated in vivo by either [ PSI +]-S7 or [ PSI +]-W2 prion variant, were isolated as described in the “PK resistance analysis of Sup35NM-GFP amyloid fibrils” chapter. The final samples were diluted 20-fold in TAE buffer with a designated urea concentration, and incubated for 8 h at 37°C without agitation. After that, the Laemmli sample buffer was added and samples were subjected to a standard SDS-PAGE procedure followed by Western blotting with immunostaining against Sup35NM. At this stage, only the monomeric Sup35NM-GFP molecules, that were released from the polymers upon their urea-mediated denaturation, are capable of entering the polyacrylamide gel. As a measure of total Sup35NM-GFP amount in polymers, similar samples without urea were boiled for 5 min, leading to a total conversion of polymers into the monomers. We used a set of urea concentrations (i.e., 0M, 4M, 6M and 8M) and found that both [ PSI +]-S7 and [ PSI +]-W2 polymers started to dissolve only in 8M urea solution, and the dissolution was much more prominent in the case of [ PSI +]-S7 polymers (Fig. S1C). Further, we made a time-lapse experiment, where the aliquots of either [ PSI +]-S7- or [ PSI +]-W2-templated fibrils were incubated at 37°C for 3–72 hours, and then analyzed as described above. The proportion of polymers dissolved in urea was calculated as a ratio of the amount of monomeric protein released in 8 M urea samples, to the amount of monomeric protein released in the boiled sample w/o urea. The experiment was performed in triplicates, and mean ± SEM values were calculated for both types of fibrils at each time point. Chaperones binding assay For this experiment, we used isogenic cells with either [ PSI+ ]-S7, [ PSI+ ]-W2 or [ psi- ] prion status, grown to a logarithmic phase. Cells were lysed using a standard lysis procedure with glass beads in 100 µl of TAE buffer supplemented with 10 mM PMSF, 1M NaN 3 and 0.01 mg/mL RNAseA. NaN 3 and RNAseA were added to disrupt large polyribosome complexes which can include monomeric Sup35 in the [ psi -] cells and, thus, to prevent monomeric Sup35 sedimentation upon the centrifugation on the later stages. Lysates were transferred into separate tubes, incubated for 10 min at 15°C, diluted to 1000 µl by TAE buffer, and spun at 6000 g for 2 min to remove unlysed cells and cell debris. 600 µl of supernatants were transferred into separate tubes and centrifuged at 20000 g for 15 min, 4°C, to precipitate prion aggregates with bound Hsp104. Supernatants were removed, 50 µl of 1X SDS-PAGE loading buffer was added, and the pellets were resuspended by vigorous pipetting. Samples were then boiled for 5 min (to dissolve Sup35 amyloid fibrils into monomers), cooled, centrifuged for 1 min at 20000 g, and then 5 µl aliquots were subjected to a standard SDS-PAGE procedure followed by immunoblotting. Each sample was immunostained by the primary polyclonal antibodies against either Sup35NM or Hsp104. The relative amounts of both Sup35 and Hsp104 proteins in each sample (in the arbitrary units) were determined by the densitometry of protein bands, and the relative Hsp104 binding to Sup35 prion aggregates was assessed in arbitrary units as: B X = (N X [ PSI+ ]−V (Hsp104) - N Mean [ psi− ] (Hsp104))/(N X [ PSI+ ]−V (Sup35) - N Mean [ psi− ] (Sup35)) where B X is a relative Hsp104 binding for a given replicate of given [ PSI+ ] variant (“ [PSI+]-V ”), N X [ PSI +]−V (Hsp104) is a Hsp104 band intensity for a given replicate of given [ PSI+ ] variant, N Mean [ psi −] (Hsp104) is a mean Hsp104 band intensity for [ psi- ] cells, N X [ PSI +]−V (Sup35) is a Sup35 band intensity for a given replicate of given [ PSI+ ] variant, and N Mean [psi−] (Sup35) is a mean Sup35 band intensity for [ psi- ] cells. The experiment was performed in four biological replicates for each cell type, and mean + SEM values were calculated for [ PSI +]-S7 and [ PSI +]-W2 prion variants. Importantly, only barely detectable Hsp104 amounts were found in the pellet of [ psi -] cells, indicating that almost all Hsp104 molecules detected in the pellets of [ PSI +] cells sedimented due to the binding to Sup35 prion aggregates. Structure-based energy calculations To calculate the free energy of unfolding (ΔG Unf ), and stability change upon mutations (i.e., the difference in free energy of unfolding between the fibrils formed by the mutated and wild-type Sup35 protein), we used FoldX 5.1 plugin in the YASARA program. ΔG Unf was calculated (default temperature = 298 K, default ionic strength = 0.05 M) using the equation: ΔG Unf = W vdw ⋅ΔG vdw + W solvH ⋅ΔG solvH + W solvP ⋅ΔG solvP + ΔG wb + ΔG hbond + ΔG el + ΔG Kon + W mc ⋅T⋅ΔS mc + + W sc ⋅T⋅ΔS sc where ΔG vdw is the sum of the van der Waals contributions of all atoms with respect to the same interactions with the solvent; ΔG solvH and ΔG solvP are the differences in solvation energy for apolar and polar groups respectively when these change from the unfolded to the folded state; ΔG hbond is the free energy difference between the formation of an intra-molecular hydrogen bond compared to inter-molecular hydrogen-bond formation (with solvent); ΔG wb is the extra stabilising free energy provided by a water molecule making more than one hydrogen bond to the protein (water bridges) that cannot be taken into account with non-explicit solvent approximations; ΔG el is the electrostatic contribution of charged groups, including the helix dipole; ΔS mc is the entropy cost of fixing the backbone in the folded state; this term is dependent on the intrinsic tendency of a particular amino acid to adopt certain dihedral angles; ΔS sc the entropic cost of fixing a side chain in a particular conformation. For a comparison with the Sup35 amyloid core structure solved in this work, the ΔG Unf values were also calculated for a set of manually selected amyloid structures representing different, partially overlapping physicochemical and biological groups: prions, pathological amyloids, functional amyloids, amyloid structures enriched in polar residues, and amyloid structures enriched in hydrophobic residues. To obtain more consistent results, all types of fibrils were standardized by the length (to be 5-mers), and the calculated ΔG Unf values were normalized for a single layer of the fibril. The solvation energy (ΔG Solv ) values were calculated using the Amyloid Illustrator package 17 , 48 (default pH = 7). As in case of ΔG Unf , the solvation energy was normalized for a single layer and compared with the values calculated for a set of published amyloid structures representing different biological and physicochemical groups. To calculate the change in ΔG Unf upon mutations (stability change, ΔΔG), we used the model of [ PSI +]-S7 fibril comprising 5 protomers, and introduced the corresponding residue substitution in all of these protomers (using the “Mutate multiple residues” feature in FoldX5.1; T = 298K, ionic strength = 0.05M, pH = 7, VdW design = 2). ΔΔG was determined as a difference between the ΔG Unf values of mutated and WT molecules. The resulting values were normalized for a single layer of the fibril. The positive values correspond to decreased structure stability, and negative values correspond to increased structure stability. Image representation Image representations of reconstructed densities and atomic models were created with UCSF Chimera 69 , ChimeraX 70 and the Amyloid Illustrator package 17 . To make an alignment between [ PSI +]-S7 (this work), Sc4 (PDB ID: 9XBN) and Sc37 (PDB ID: 9XBO) structures, we used the Pairwise Structure Alignment tool in the www.rcsb.org web-site (TM-align algorithm). Data availability Atomic coordinates of [ PSI+ ]-S7 structure have been deposited in the Protein Data Bank under accession number 21BQ. The corresponding cryo-EM density map has been deposited in the Electron Microscopy Data Bank under accession code EMD-67558. Declarations Acknowledgments We thank Resource Center of Probe and Electron Microscopy of National Research Center “Kurchatov Institute” (http://rc.nrcki.ru/pages/main/nanozond/facilities/12604/index.shtml). This work has been carried out using computing resources of the federal collective usage center Complex for Simulation and Data Processing for Mega-science Facilities at National Research Center “Kurchatov Institute” (http://ckp.nrcki.ru). MALDI mass spectrometry and DNA sequencing were performed by the Shared Access Equipment Centre “Industrial Biotechnology” of the FRC “Fundamentals of Biotechnology” (RAS). We thank Valery Urakov for the spheroplast prion transformation and Andrey Moiseenko (Biology Faculty of Moscow State University, Moscow) for the help with negative staining electron microscopy experiments. Funding Statement This research was funded by the Russian Science Foundation, grant #23-74-00062 (A.A.D, A.D.B, V.V.K), and in part by the Ministry of Science and Higher Education of the Russian Federation (O.M.V, K.M.B, V.O.P). Author contributions Conceptualization: A.A.D, V.V.K; methodology: A.A.D, Y.M.C., A.D.B., V.V.K, K.M.B; acquisition of data: A.A.D, Y.M.C., A.D.B., O.V.M, T.N.B.; analysis of data: Y.M.C., A.A.D, A.D.B, V.V.K, T.N.B.; writing — original draft: A.A.D; writing — review and editing: V.V.K, A.A.D, Y.M.C., A.D.B., K.M.B., V.O.P; visualization: A.A.D, Y.M.C.; supervision: V.V.K, A.A.D, K.M.B; funding acquisition: V.V.K, V.O.P. All authors have read and agreed to the published version of the manuscript. References Stefani M, Dobson CM (2003) Protein aggregation and aggregate toxicity: new insights into protein folding, misfolding diseases and biological evolution. J Mol Med 81:678–699 Eisenberg D, Jucker M (2012) The Amyloid State of Proteins in Human Diseases. Cell 148:1188–1203 Sergeeva AV, Galkin AP (2020) Functional amyloids of eukaryotes: criteria, classification, and biological significance. Curr Genet 66:849–866 Otzen D, Riek R (2019) Functional Amyloids. Cold Spring Harb Perspect Biol 11:a033860 Galkin AP, Sysoev EI, Valina AA (2023) Amyloids and prions in the light of evolution. Curr Genet 69:189–202 Wickner RB, Masison DC, Edskes HK (1995) [PSI] and [URE3] as yeast prions. Yeast 11:1671–1685 Paushkin SV, Kushnirov VV, Smirnov VN, Ter-Avanesyan MD (1996) Propagation of the yeast prion-like [psi+] determinant is mediated by oligomerization of the SUP35-encoded polypeptide chain release factor. EMBO J 15:3127–3134 Patino MM, Liu JJ, Glover JR, Lindquist S (1996) Support for the prion hypothesis for inheritance of a phenotypic trait in yeast. Science 273:622–626 Liebman SW, Chernoff YO (2012) Prions Yeast Genetics 191:1041–1072 Shen CH et al (2024) Exposed Hsp70-binding site impacts yeast Sup35 prion disaggregation and propagation. Proc. Natl. Acad. Sci. U.S.A. 121, e2318162121 Collinge J, Clarke AR (2007) A General Model of Prion Strains and Their Pathogenicity. Science 318:930–936 Collinge J (2016) Mammalian prions and their wider relevance in neurodegenerative diseases. Nature 539:217–226 Jucker M, Walker LC (2013) Self-propagation of pathogenic protein aggregates in neurodegenerative diseases. Nature 501:45–51 Goedert M (2015) Alzheimer’s and Parkinson’s diseases: The prion concept in relation to assembled Aβ, tau, and α-synuclein. Science 349:1255555 Goedert M, Masuda-Suzukake M, Falcon B (2017) Like prions: the propagation of aggregated tau and α-synuclein in neurodegeneration. Brain 140:266–278 Qiang W, Yau W-M, Lu J-X, Collinge J, Tycko R (2017) Structural variation in amyloid-β fibrils from Alzheimer’s disease clinical subtypes. Nature 541:217–221 Sawaya MR, Hughes MP, Rodriguez JA, Riek R, Eisenberg DS (2021) The expanding amyloid family: Structure, stability, function, and pathogenesis. Cell 184:4857–4873 Scheres SHW, Ryskeldi-Falcon B, Goedert M (2023) Molecular pathology of neurodegenerative diseases by cryo-EM of amyloids. Nature 621:701–710 Khurana V, Lindquist S (2010) Modelling neurodegeneration in Saccharomyces cerevisiae: why cook with baker’s yeast? Nat Rev Neurosci 11:436–449 Kochneva-Pervukhova NV et al (2001) [PSI+] prion generation in yeast: characterization of the strain difference. Yeast 18:489–497 Uptain SM, Sawicki GJ, Caughey B, Lindquist S (2001) Strains of [PSI(+)] are distinguished by their efficiencies of prion-mediated conformational conversion. EMBO J 20:6236–6245 Dergalev AA, Alexandrov AI, Ivannikov RI, Ter-Avanesyan MD, Kushnirov VV (2019) Yeast Sup35 Prion Structure: Two Types, Four Parts, Many Variants. Int J Mol Sci 20 Huang Y-W, King C-Y (2019) A complete catalog of wild-type Sup35 prion variants and their protein-only propagation. Curr Genet. 10.1007/s00294-019-01003-8 Chernoff Y, Lindquist S, Ono B, Inge-Vechtomov S, Liebman S (1995) Role of the chaperone protein Hsp104 in propagation of the yeast prion-like factor [psi+]. Science 268:880–884 Tipton KA, Verges KJ, Weissman JS (2008) In vivo monitoring of the prion replication cycle reveals a critical role for Sis1 in delivering substrates to Hsp104. Mol Cell 32:584–591 Shorter J, Lindquist S (2008) Hsp104, Hsp70 and Hsp40 interplay regulates formation, growth and elimination of Sup35 prions. EMBO J 27:2712–2724 Tanaka M, Chien P, Naber N, Cooke R, Weissman JS (2004) Conformational variations in an infectious protein determine prion strain differences. Nature 428:323–328 Tanaka M, Collins SR, Toyama BH, Weissman JS (2006) The physical basis of how prion conformations determine strain phenotypes. Nature 442:585–589 Toyama BH, Kelly MJS, Gross JD, Weissman JS (2007) The structural basis of yeast prion strain variants. Nature 449:233–237 Ohhashi Y et al (2018) Molecular basis for diversification of yeast prion strain conformation. Proc Natl Acad Sci U S A 115:2389–2394 Alexandrov AI, Polyanskaya AB, Serpionov GV, Ter-Avanesyan MD, Kushnirov VV (2012) The effects of amino acid composition of glutamine-rich domains on amyloid formation and fragmentation. PLoS ONE 7:e46458 Tanaka M et al (2025) How Sup35 monomer conformation and amyloid fibril polymorphism determine yeast strain phenotypes. Preprint at. https://doi.org/10.21203/rs.3.rs-7945345/v1 Kryndushkin DS, Alexandrov IM, Ter-Avanesyan MD, Kushnirov VV (2003) Yeast [ PSI + ] Prion Aggregates Are Formed by Small Sup35 Polymers Fragmented by Hsp104. J Biol Chem 278:49636–49643 Cox B, Ness F, Tuite M (2003) Analysis of the generation and segregation of propagons: entities that propagate the [PSI+] prion in yeast. Genetics 165:23–33 Bagriantsev SN, Gracheva EO, Richmond JE, Liebman SW (2008) Variant-specific [PSI+] infection is transmitted by Sup35 polymers within [PSI+] aggregates with heterogeneous protein composition. Mol Biol Cell 19:2433–2443 Derkatch IL, Bradley ME, Zhou P, Liebman SW (1999) The PNM2 mutation in the prion protein domain of SUP35 has distinct effects on different variants of the [PSI+] prion in yeast. Curr Genet 35:59–67 Verges KJ, Smith MH, Toyama BH, Weissman JS (2011) Strain conformation, primary structure and the propagation of the yeast prion [PSI+]. Nat Struct Mol Biol 18:493–499 Pei F, DiSalvo S, Sindi SS, Serio TR (2017) A dominant-negative mutant inhibits multiple prion variants through a common mechanism. PLoS Genet 13:e1007085 Dergalev AA, Urakov VN, Agaphonov MO, Alexandrov AI, Kushnirov VV (2021) Dangerous Stops: Nonsense Mutations Can Dramatically Increase Frequency of Prion Conversion. IJMS 22:1542 DePace AH, Santoso A, Hillner P, Weissman JS (1998) A Critical Role for Amino-Terminal Glutamine/Asparagine Repeats in the Formation and Propagation of a Yeast Prion. Cell 93:1241–1252 Lawrence MC, Colman PM (1993) Shape Complementarity at Protein/Protein Interfaces. J Mol Biol 234:946–950 Sawaya MR et al (2007) Atomic structures of amyloid cross-beta spines reveal varied steric zippers. Nature 447:453–457 Schymkowitz J et al (2005) The FoldX web server: an online force field. Nucleic Acids Res 33:W382–W388 Lu J et al (2024) Cryo-EM structures of the D290V mutant of the hnRNPA2 low-complexity domain suggests how D290V affects phase separation and aggregation. J Biol Chem 300:105531 Cao Q, Boyer DR, Sawaya MR, Ge P, Eisenberg DS (2019) Cryo-EM structures of four polymorphic TDP-43 amyloid cores. Nat Struct Mol Biol 26:619–627 Rosenberg GM et al (2023) Fibril structures of TFG protein mutants validate the identification of TFG as a disease-related amyloid protein by the IMPAcT method. PNAS Nexus 2:pgad402 De Ibáñez A et al (2022) Molecular interactions of FG nucleoporin repeats at high resolution. Nat Chem 14:1278–1285 Eisenberg D, McLachlan AD (1986) Solvation energy in protein folding and binding. Nature 319:199–203 Hervas R et al (2020) Cryo-EM structure of a neuronal functional amyloid implicated in memory persistence in Drosophila . Science 367:1230–1234 Marchante R, Rowe M, Zenthon J, Howard MJ, Tuite MF (2013) Structural Definition Is Important for the Propagation of the Yeast [PSI+] Prion. Mol Cell 50:675–685 King C-Y (2001) Supporting the structural basis of prion strains: induction and identification of [PSI] variants. J Mol Biol 307:1247–1260 Parham SN (2001) Oligopeptide repeats in the yeast protein Sup35p stabilize intermolecular prion interactions. EMBO J 20:2111–2119 Shkundina IS, Kushnirov VV, Tuite MF, Ter-Avanesyan MD (2006) The Role of the N-Terminal Oligopeptide Repeats of the Yeast Sup35 Prion Protein in Propagation and Transmission of Prion Variants. Genetics 172:827–835 Huang Y, Kushnirov VV, King C (2021) Mutable yeast prion variants are stabilized by a defective Hsp104 chaperone. Mol Microbiol 115:774–788 Kushnirov VV, Ter-Avanesyan MD (1998) Structure and Replication of Yeast Prions. Cell 94:13–16 Shewmaker F, Wickner RB, Tycko R (2006) Amyloid of the prion domain of Sup35p has an in-register parallel beta-sheet structure. Proc. Natl. Acad. Sci. U.S.A. 103, 19754–19759 Shewmaker F, Kryndushkin D, Chen B, Tycko R, Wickner RB (2009) Two Prion Variants of Sup35p Have In-Register Parallel β-Sheet Structures, Independent of Hydration. Biochemistry 48:5074–5082 Gorkovskiy A, Thurber KR, Tycko R, Wickner RB (2014) Locating folds of the in-register parallel β-sheet of the Sup35p prion domain infectious amyloid. Proc. Natl. Acad. Sci. U.S.A. 111, E4615-4622 Krishnan R, Lindquist SL (2005) Structural insights into a yeast prion illuminate nucleation and strain diversity. Nature 435:765–772 Roberts BE et al (2009) A synergistic small-molecule combination directly eradicates diverse prion strain structures. Nat Chem Biol 5:936–946 King C-Y, Diaz-Avalos R (2004) Protein-only transmission of three yeast prion strains. Nature 428:319–323 Galliamov AA, Urakov VN, Dergalev AA, Kushnirov VV (2025) On the Significance of the Terminal Location of Prion-Forming Regions of Yeast Proteins. IJMS 26, 1637 Ansari S et al (2024) In cell NMR reveals cells selectively amplify and structurally remodel amyloid fibrils. Preprint at. https://doi.org/10.1101/2024.09.09.612142 Galliamov AA, Malukhina AD, Kushnirov VV (2024) Mapping of Prion Structures in the Yeast Rnq1. IJMS 25, 3397 Betzel C et al (2001) Structure of a Serine Protease Proteinase K from Tritirachium album limber at 0.98 Å Resolution. Biochemistry 40:3080–3088 Gietz RD, Schiestl RH (2007) High-efficiency yeast transformation using the LiAc/SS carrier DNA/PEG method. Nat Protoc 2:31–34 Saibil HR et al (2012) Heritable yeast prions have a highly organized three-dimensional architecture with interfiber structures. Proc. Natl. Acad. Sci. U.S.A. 109, 14906–14911 Burtseva AD et al (2023) Electron Microscopy Study of the Structure of the Sup35 Prion from Yeast Saccharomyces cerevisiae. Crystallogr Rep 68:872–878 Pettersen EF et al (2004) UCSF Chimera—A visualization system for exploratory research and analysis. J Comput Chem 25:1605–1612 Pettersen EF et al (2021) UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci 30:70–82 Additional Declarations The authors declare no competing interests. Supplementary Files Supplementarytext.docx Sup35CryoEMstructureFiguresSupplementarypreprint.pdf Cite Share Download PDF Status: Posted 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-8764517","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":584312541,"identity":"527eefbb-58ef-4585-8e39-afaad8d65bc6","order_by":0,"name":"Dergalev Alexander A.","email":"data:image/png;base64,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","orcid":"","institution":"Bach Institute of Biochemistry, Federal Research Center of Biotechnology of the RAS","correspondingAuthor":true,"prefix":"","firstName":"Dergalev","middleName":"Alexander","lastName":"A.","suffix":""},{"id":584342193,"identity":"a1c884a0-3364-4978-a5d8-b81f4cd9c29f","order_by":1,"name":"Chesnokov Yuri M.","email":"","orcid":"","institution":"National Research Center “Kurchatov Institute”","correspondingAuthor":false,"prefix":"","firstName":"Chesnokov","middleName":"Yuri","lastName":"M.","suffix":""},{"id":584342318,"identity":"f032bf1b-26f9-4054-95d9-3c02ca490cbb","order_by":2,"name":"Burtseva Anna D.","email":"","orcid":"","institution":"Moscow Center for Advanced Studies","correspondingAuthor":false,"prefix":"","firstName":"Burtseva","middleName":"Anna","lastName":"D.","suffix":""},{"id":584351337,"identity":"9ccadc13-f6b1-4cfc-bc43-bff8f32db6f5","order_by":3,"name":"Mitkevich Olga V.","email":"","orcid":"","institution":"Bach Institute of Biochemistry, Federal Research Center of Biotechnology of the RAS","correspondingAuthor":false,"prefix":"","firstName":"Mitkevich","middleName":"Olga","lastName":"V.","suffix":""},{"id":584351338,"identity":"424d3c61-dbc2-4fb9-9f8a-eeb2db280023","order_by":4,"name":"Baymukhametov Timur N.","email":"","orcid":"","institution":"National Research Center “Kurchatov Institute”","correspondingAuthor":false,"prefix":"","firstName":"Baymukhametov","middleName":"Timur","lastName":"N.","suffix":""},{"id":584352687,"identity":"d52e7e57-96d7-4483-9fd8-4b336598f851","order_by":5,"name":"Popov Vladimir O.","email":"","orcid":"","institution":"Bach Institute of Biochemistry, Federal Research Center of Biotechnology of the RAS","correspondingAuthor":false,"prefix":"","firstName":"Popov","middleName":"Vladimir","lastName":"O.","suffix":""},{"id":584352688,"identity":"635dba42-d608-40ab-ac30-c030d8eeb12e","order_by":6,"name":"Boyko Konstantin M.","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAArklEQVRIiWNgGAWjYBACCQYexgcJDAd4SNLCbECyFjYJBoYDJDhMctrZYxUPd9yRYZA+/PgD457DhLVIS+el3Ug884yHgS/NTILhGRFa5KRzzG4kth3mYeBhMAM6kEgtBRAt7J8/EKVFGqiFAaKFx0CCKC2Ss/OSJRLbnvGw8fCUSSQcSCesReJ27sGPP9vu2PPzsG/+8OGANWEtcMAGIhJI0DAKRsEoGAWjAA8AACSnNCPp8iiMAAAAAElFTkSuQmCC","orcid":"","institution":"Bach Institute of Biochemistry, Federal Research Center of Biotechnology of the RAS","correspondingAuthor":true,"prefix":"","firstName":"Boyko","middleName":"Konstantin","lastName":"M.","suffix":""},{"id":584352689,"identity":"dc51b4fa-ffc6-4dba-a41b-3663a1d9097f","order_by":7,"name":"Kushnirov Vitaly V.","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1ElEQVRIiWNgGAWjYFACNgYGxgYGHiCL8QGQ4OEjRQuzAUgLG7FawCwJKB8/MLiRlib5c8cdGfP2HrPKrzl2MmwMzA8f3cCv5Zg075lnPDJnzpjdlt2WDHQYm7FxDh4tkj3H26QZ2w7zSEikpd2W3MYM1MLDJk1Ii+RPqJZiyW31hLXws7cdk+AFa0k+xvhx22GitCRbg7XwHD4szbjtOA8bMwG/ABUY3gQ6zF6CvbHx489t1fb87M0PH+PTggKYecAkscpBgPEHKapHwSgYBaNgxAAA5yk/d+gvnK8AAAAASUVORK5CYII=","orcid":"","institution":"Bach Institute of Biochemistry, Federal Research Center of Biotechnology of the RAS","correspondingAuthor":true,"prefix":"","firstName":"Kushnirov","middleName":"Vitaly","lastName":"V.","suffix":""}],"badges":[],"createdAt":"2026-02-02 11:51:44","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8764517/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8764517/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101744614,"identity":"90ed3103-c6ee-496d-898f-e0024eb2a6c1","added_by":"auto","created_at":"2026-02-03 09:00:06","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":328176,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProperties of strong and weak [\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePSI\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e+] prion variants used in this work.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003ePhenotype of the strong and weak [\u003cem\u003ePSI\u003c/em\u003e+] variants in the \u003cem\u003eade1-14 \u003c/em\u003eyeast strain.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB. \u003c/strong\u003eComparison of the Sup35 polymer size by the SDD-AGE and immunoblotting.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC. \u003c/strong\u003eComparison of the proportion of soluble Sup35 in the cells carrying strong and weak [\u003cem\u003ePSI\u003c/em\u003e+] variants, determined as a proportion of Sup35 molecules capable of entering a 10% polyacrylamide gel without boiling.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD-F\u003c/strong\u003e. Comparison of resistance to heat (thermal stability assay, \u003cstrong\u003eD\u003c/strong\u003e), urea (\u003cstrong\u003eE\u003c/strong\u003e) and mechanical shearing via ultrasonication (\u003cstrong\u003eF\u003c/strong\u003e) for [PSI+]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 prion fibrils. In \u003cstrong\u003eD\u003c/strong\u003e, the fractions of Sup35 prion fibrils, which resisted the indicated temperatures, were determined by SDD-AGE. In \u003cstrong\u003eE\u003c/strong\u003e, purified Sup35NM-GFP fibrils were incubated for the indicated time in 8 M urea, and the proportion of released monomeric Sup35NM-GFP was determined by SDS-PAGE, using a sample boiled without urea as a measure of the total Sup35NM-GFP amount in fibrils. In \u003cstrong\u003eF\u003c/strong\u003e, equal aliquots of Sup35NM-GFP fibrils were ultrasonicated for different durations, and the change in fibril size distribution was assessed by SDD-AGE. In all cases, electrophoresis was followed by immunoblotting and densitometry of blots.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eG. \u003c/strong\u003eHsp104 binding to strong- and weak-type Sup35 prion aggregates assessed via their co-sedimentation from cell lysates. The relative quantity of Hsp104 and aggregated Sup35 in pellets was assessed by Western blotting, using [\u003cem\u003epsi\u003c/em\u003e-] cells as a negative control.\u003c/p\u003e\n\u003cp\u003eP-values were calculated using two-tailed Student’s t-test.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/2030e79f5eba4ee7f8fad411.jpg"},{"id":101754136,"identity":"4c425986-4e46-4e0a-a0e3-ca5c26f06541","added_by":"auto","created_at":"2026-02-03 10:41:43","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":777162,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDissecting the roles of Core1A and Core1B in the stability and phenotype of strong and weak [\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePSI+\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e] variants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eSchematic representation of the Sup35 N-domain showing its Q/N-rich and oligopeptide repeats (R1-R4) regions, location of the fully PK-resistant Core1A and partly PK-resistant Core1B, and the studied anti-prion mutations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB. \u003c/strong\u003eNonsense-suppressor and \u003cem\u003eAde+\u003c/em\u003e phenotypes of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells with the \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003egene shuffled for either \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eG58D\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ24R \u003c/em\u003e\u003c/sup\u003eor \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eS17R\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003emutant alleles.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC.\u003c/strong\u003e SDS-resistant Sup35 polymers detected by SDD-AGE in the lysates of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells where \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003ewas shuffled for the indicated mutant variants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD. \u003c/strong\u003eProportions of soluble Sup35 and Sup35\u003csup\u003eG58D \u003c/sup\u003ein [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells, estimated as by the Sup35 ability to enter a 10% polyacrylamide gel without boiling.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eE. \u003c/strong\u003eMapping of PK-resistant cores of prion fibrils isolated from [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells and composed of either Sup35NM\u003csup\u003eWT\u003c/sup\u003e-GFP or Sup35NM\u003csup\u003eG58D\u003c/sup\u003e-GFP proteins.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eF-G. \u003c/strong\u003eComparison of the resistance to heat (thermal stability assay, \u003cstrong\u003eF\u003c/strong\u003e) and mechanical shearing by ultrasonication (\u003cstrong\u003eG\u003c/strong\u003e) for strong- and weak-type prion polymers consisting of either Sup35NM\u003csup\u003eWT\u003c/sup\u003e-GFP or Sup35NM\u003csup\u003eG58D\u003c/sup\u003e-GFP proteins.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eH.\u003c/strong\u003e Comparison of prion mitotic stability for [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells with the \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003eallele replaced for the \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eG58D \u003c/em\u003e\u003c/sup\u003eor \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ24R \u003c/em\u003e\u003c/sup\u003ealleles.\u003cbr\u003e\nP-values were calculated using two-tailed Student’s t-test\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/70bc32358592fcb0cd3c860a.jpg"},{"id":101754138,"identity":"e3d08c41-2d07-4fd2-860f-241a6fb66462","added_by":"auto","created_at":"2026-02-03 10:41:44","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":607801,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCryo-EM structure of the [\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePSI+\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e]-S7 prion fibril core.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eCross-sectional view of the 3D reconstruction of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid core structure.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB-C.\u003c/strong\u003e Rendered 3D model of the ordered part of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid fibril (\u003cstrong\u003eB\u003c/strong\u003e), and the atomic model of a single protomer superimposed on the sharpened cryo-EM density map (\u003cstrong\u003eC\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD.\u003c/strong\u003e The topology of the [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid fold, showing the location of β-strands. The sidechains are color-coded based on their physiochemical properties.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eE.\u003c/strong\u003e Atomic model of the [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid core showing the “inner” and “outer” steric zipper elements.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eF. \u003c/strong\u003eClose-up of the parts of “inner” and “outer” steric zippers showing their “staggered” organization along the Z-axis of the fibril.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eG. \u003c/strong\u003eClose-up of several 𝛑-columns formed by tyrosine residues (green arrows) and amide ladders formed by glutamine residues (violet arrows) stabilizing the core structure along the axis of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 prion fibrils.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eH-I.\u003c/strong\u003e Comparison of the PK-protected region and cryo-EM-solved ordered structure of Core1. Dotted lines indicate the main stop sites for proteinase K digestion of Core1B. The “inner” and “outer” steric zipper zones are highlighted in magenta and blue, respectively. The major PK-resistant peptides observed in the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 preparation are shown below.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/e5788a767ab13bd06e90929a.jpg"},{"id":101744613,"identity":"c32c968c-e04f-4b1c-9a5b-e75425e048c7","added_by":"auto","created_at":"2026-02-03 09:00:06","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":351730,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThermodynamic profiling of the [\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ePSI+\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e]-S7 core structure and the effects of anti-prion mutations.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eGibbs free energy of unfolding and solvation energy of Sup35/[\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion structure (red arrow) in comparison with that of several representatives of different amyloid classes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB. \u003c/strong\u003eAmyloid core structure with known anti-prion mutations\u003csup\u003e40,50,51\u003c/sup\u003e. Most of them represent uncharged-to-charged residues substitutions in the zipper-forming regions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC. \u003c/strong\u003ePredicted change in the free energy of unfolding (ΔΔG) calculated by FoldX software for the anti-prion Sup35 mutations in comparison with the wild type structure. In contrast to known anti-prion mutations, manually selected substitutions (uncharged-to-charged residues) located outside of the zipper-forming regions have only a minor impact on prion stability. P-value was calculated using two-tailed Student’s t-test.\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/5089c69a1c131ab5790027c2.jpg"},{"id":101744619,"identity":"0297580b-d5ce-411d-b2a0-d9a5a555e0e4","added_by":"auto","created_at":"2026-02-03 09:00:06","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":360423,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure 6. Comparison of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eex vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e Sup35 prion structure with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ein vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e formed Sup35 amyloids\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eStructure of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 \u003cem\u003eex vivo\u003c/em\u003e prion fibrils in comparison with the Sc4 (PDB ID: 9XBN) and Sc37 (PDB ID: 9XBO) amyloids formed \u003cem\u003ein vitro\u003c/em\u003e\u003csup\u003e27,28\u003c/sup\u003e. Steric zipper elements are highlighted in grey.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB-C. \u003c/strong\u003eStructural alignment of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 with Sc4 (\u003cstrong\u003eB\u003c/strong\u003e) and Sc37 (\u003cstrong\u003eC\u003c/strong\u003e) structures, with the close-up of some regions fully or partially similar between [\u003cem\u003ePSI+\u003c/em\u003e]-S7 with Sc4 structures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eD.\u003c/strong\u003e Comparison of the free energy of unfolding (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e) and solvation energy (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e) between [\u003cem\u003ePSI+\u003c/em\u003e]-S7, Sc4 and Sc37 structures.\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/1d15a1afa2d5277b7563253c.jpg"},{"id":101744620,"identity":"f10f2bed-4dda-40bd-b96b-8d65cd59a09d","added_by":"auto","created_at":"2026-02-03 09:00:06","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":311002,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA-B. \u003c/strong\u003eTwo possible explanations of relation between the structures of \u003cem\u003ein vitro\u003c/em\u003e formed Sc4 amyloid and \u003cem\u003ein vivo\u003c/em\u003e [\u003cem\u003ePSI+\u003c/em\u003e]-strong prion variant. First one (\u003cstrong\u003eA\u003c/strong\u003e) implies structural rearrangement of Sc4 amyloid by the extension of existing cross-β fold on some unfolded parts of Sup35 molecule (the N36-Y46 and G54-N64 stretches), and the second one (\u003cstrong\u003eB\u003c/strong\u003e) implies positive selection of rare, but the most fitted Sc4 polymorphs inside the yeast cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC. \u003c/strong\u003eProposed mechanism of the anti-prion action of the G58D and Q24R substitutions with respect to the strong [\u003cem\u003ePSI+\u003c/em\u003e] variant structure resolved in this work.\u003c/p\u003e","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/187622220e5bf09c78c1a9f4.jpg"},{"id":101755915,"identity":"eebb304d-3e0a-4ade-afd7-7e677b0f42b6","added_by":"auto","created_at":"2026-02-03 10:55:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4438772,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/850def6b-11be-450e-9949-07667de80cde.pdf"},{"id":101753903,"identity":"8010c666-5ca7-4d16-a7e8-171d363948ef","added_by":"auto","created_at":"2026-02-03 10:41:09","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":25013,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarytext.docx","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/eef3597fae402abce08a8367.docx"},{"id":101754487,"identity":"c74bc653-4d52-4521-9f80-996cbab7f6d1","added_by":"auto","created_at":"2026-02-03 10:42:43","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3479488,"visible":true,"origin":"","legend":"","description":"","filename":"Sup35CryoEMstructureFiguresSupplementarypreprint.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8764517/v1/4ec1f649604d3671a67266f9.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eCryo-EM structure of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eex vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e Sup35 yeast prion and the factors defining prion phenotype\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAmyloids are filamentous protein polymers, which contain highly ordered core with cross-ꞵ structure and may also include structured and unstructured parts. Amyloids are primarily known as pathological agents related to some incurable degenerative diseases of mammals\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e \u0026ndash; but, as it is now known, they are widespread across all kingdoms of life, and often play an important role in the normal physiology of various organisms\u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Amyloids, which can either spontaneously infect multicellular organisms, or stably propagate in microorganisms across many generations, are called prions. The [\u003cem\u003ePSI+\u003c/em\u003e] prion of yeast \u003cem\u003eS. cerevisiae\u003c/em\u003e represents heritable amyloid of the eRF3 translation termination factor, also known as the Sup35 protein\u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. When Sup35 switches into the prion form, translation termination efficiency decreases. Thus, the [\u003cem\u003ePSI+\u003c/em\u003e] prion is an epigenetic element with a loss-of-function phenotype\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe Sup35 protein is composed of three domains. The N-terminal domain (N, residues 2-123) is mainly responsible for the prion properties and rich in glutamine and asparagine (QN). The middle domain (M, residues 124\u0026ndash;253) is rich in glutamic acid and lysine and includes the major target for Ssa1/2 chaperones (residues 143\u0026ndash;164)\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. The C-terminal domain (C, residues 254\u0026ndash;685) performs an essential function in the translation termination.\u003c/p\u003e \u003cp\u003eMany amyloids exhibit structural polymorphism \u0026ndash; i.e., the ability of the same amyloid-forming protein to acquire many different cross-ꞵ structures. Different amyloid polymorphs of α-synuclein, tau protein, Aꞵ peptide, PrP and other human proteins were shown to be associated with different types of respective neurodegenerative disorders\u003csup\u003e\u003cspan additionalcitationids=\"CR12 CR13 CR14 CR15\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Although many such pathology-related amyloid structures are now established\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, the mechanisms relating these structures to pathologic phenotypes remain poorly understood due to the high complexity of multicellular organisms.\u003c/p\u003e \u003cp\u003eIn contrast, baker\u0026rsquo;s yeast is a simple unicellular organism, where the phenotypic manifestations of prions and their relation to prion structures can be observed and studied more easily\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. The [\u003cem\u003ePSI+\u003c/em\u003e] prion phenotype is usually detected using nonsense mutations in adenine biosynthesis pathway genes, which cause accumulation of red intermediate of adenine biosynthesis. The Sup35 prion state reduces the efficiency of translation termination allowing partial readthrough of mutant nonsense codons, and reverting colony color from red to some shades of pink or white, depending on the extent of Sup35 inactivation (nonsense suppressor phenotype).\u003c/p\u003e \u003cp\u003eThe [\u003cem\u003ePSI+\u003c/em\u003e] prion also has a high number of structural variants differing by the level of nonsense suppression that are usually roughly divided into \u0026ldquo;strong\u0026rdquo; and \u0026ldquo;weak\u0026rdquo; types based on colony color phenotype\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. This criterion is not sufficiently strict, but we have shown that strong and weak variants represent two distinct classes that can be reliably distinguished by their proteinase K (PK) digestion patterns and opposite reaction to overproduction of \u003cem\u003eSUP35\u003c/em\u003e and \u003cem\u003eHSP104\u003c/em\u003e genes\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. The most systematic study of the [\u003cem\u003ePSI+\u003c/em\u003e] prion variation was able to distinguish 23 variants\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. However, only five of them, including one strong and four of weak variants, were obtained through Sup35 overproduction, while others were generated by prion passage on various Sup35 mutants.\u003c/p\u003e \u003cp\u003eIt is widely assumed that many, if not all, phenotypic features distinguishing strong and weak variants result from the more frequent fragmentation of the strong Sup35 prions, compared to weak ones by Hsp104/Ssa1/Sis1 chaperone mechanism in yeast\u003csup\u003e\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. If so, the study of structure-to-phenotype relation in case of [\u003cem\u003ePSI+\u003c/em\u003e] prion variants could be reduced to a significant extent to seeking the mechanism relating the structures of Sup35 prions and the frequency of their fragmentation in yeast.\u003c/p\u003e \u003cp\u003eCurrent understanding of the properties of Sup35 prions is largely based on studies of the amyloid fibrils spontaneously formed \u003cem\u003ein vitro\u003c/em\u003e by the Sup35NM protein lacking the C domain. The fibrils formed at 4˚C produce mainly strong [\u003cem\u003ePSI+\u003c/em\u003e] prion upon transfection into yeast cells, while those formed at 37˚C produce mainly weak [\u003cem\u003ePSI+\u003c/em\u003e]\u003csup\u003e27,28\u003c/sup\u003e. Due to this, it is commonly assumed that the former fibrils (Sc4) are structurally equivalent to the strong prion variant, while the latter (Sc37) are equivalent to the weak one. Deuterium exchange and proteinase K (PK) digestion data indicate that the Sup35 amyloid cores include amino acid residues 2\u0026ndash;46 in Sc4 and 2\u0026ndash;72 in Sc37\u003csup\u003e29,30\u003c/sup\u003e. The Sc4 fibrils also showed significantly higher fragility than Sc37\u003csup\u003e28\u003c/sup\u003e. This allowed thinking that the higher fragility of strong prion variants defines their easier and more frequent fragmentation compared to weak ones. An alternative idea was that Hsp104 is sufficiently powerful to fragment any prion, but the critical point is the ability of the chaperone machinery to recognize prion as a target, and smaller prion core of strong prions leaves longer unfolded regions for better chaperone recognition\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, PK mapping of prion cores of 26 strong and weak Sup35 prions isolated from yeast revealed that they all have a core structure comprising residues 2\u0026ndash;72. However, while residues 2\u0026ndash;32 are fully protected, the residues 33\u0026ndash;72 are protected only partly and significantly better protected in strong variants\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. This causes a paradox that by PK maps strong [\u003cem\u003ePSI+\u003c/em\u003e] are highly similar to Sc37 amyloid, rather than to Sc4, while weak [\u003cem\u003ePSI+\u003c/em\u003e] is similar to Sc4\u003csup\u003e22,30\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCurrent work, together with that of Tanaka et al.\u003csup\u003e32\u003c/sup\u003e resolves this paradox by showing the substantial structural difference of the Sup35 amyloids obtained \u003cem\u003ein vitro\u003c/em\u003e and prion polymers propagating in yeast cells. Using the cryo-EM approach together with biochemical, computational, and genetic methods, we established the 3D-structure of the N-terminal core of strong-type Sup35 amyloid fibrils, analyzed the impact of different parts of Core 1 on the overall stability of these fibrils, and an effect of different mutations earlier reported to have anti-prion action. Based on these findings, we propose a model explaining how different Sup35 structures predetermine the different frequency of fibril fragmentation \u003cem\u003ein vivo\u003c/em\u003e, and, eventually, the different severity of the [\u003cem\u003ePSI+\u003c/em\u003e] prion phenotype.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003e1. [\u003c/b\u003e \u003cb\u003ePSI+\u003c/b\u003e \u003cb\u003e]-S7 and [\u003c/b\u003e \u003cb\u003ePSI+\u003c/b\u003e \u003cb\u003e]-W2 prion fibrils differ in fragmentation due to their physical properties, rather than chaperone binding\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo study the structural basis of the difference of strong and weak [\u003cem\u003ePSI\u003c/em\u003e+] variants, we took [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 as the typical representatives of the \u0026ldquo;strong\u0026rdquo; and \u0026ldquo;weak\u0026rdquo; classes characterized in our previous work\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Consistent with earlier observations made by our and other groups on strong and weak [\u003cem\u003ePSI\u003c/em\u003e+] variants\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e, the cells bearing strong [\u003cem\u003ePSI\u003c/em\u003e+]-S7 variant had smaller size of Sup35 polymers, lower proportion of Sup35 monomers, and higher number of propagons (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA-\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC, S1A) compared to the weak [\u003cem\u003ePSI\u003c/em\u003e+]-W2 variant.\u003c/p\u003e \u003cp\u003eWe compared the mechanical properties of Sup35 prions from [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 cells and their interaction with chaperones. The [\u003cem\u003ePSI\u003c/em\u003e+]-S7 prion was less resistant to heating and high-molar urea (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD-\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE, S1C), and more fragile (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF). Assaying the amyloid/chaperone co-sedimentation, we observed that Sup35 fibrils of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 variants bind similar amounts of the Hsp104 chaperone (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), the key player in fragmentation and disaggregation of Sup35 polymers\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. This agrees with the finding that strong- and weak-type Sup35 prion polymers bind equal amounts of the Ssa1/2 chaperones\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. These results indicate that strong- and weak-type polymers represent equally attractive substrates for the chaperones performing amyloid fragmentation. Summing up, our data suggest that the fibrils stability is the primary feature that determines the frequency of their fragmentation, the balance between soluble and amyloid Sup35 protein and, eventually, the \"strength\" of the [\u003cem\u003ePSI\u003c/em\u003e+] prion phenotype.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eA.\u003c/b\u003e Phenotype of the strong and weak [\u003cem\u003ePSI\u003c/em\u003e+] variants in the \u003cem\u003eade1-14\u003c/em\u003e yeast strain.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eB.\u003c/b\u003e Comparison of the Sup35 polymer size by the SDD-AGE and immunoblotting.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eC.\u003c/b\u003e Comparison of the proportion of soluble Sup35 in the cells carrying strong and weak [\u003cem\u003ePSI\u003c/em\u003e+] variants, determined as a proportion of Sup35 molecules capable of entering a 10% polyacrylamide gel without boiling.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eD-F\u003c/b\u003e. Comparison of resistance to heat (thermal stability assay, \u003cb\u003eD\u003c/b\u003e), urea (\u003cb\u003eE\u003c/b\u003e) and mechanical shearing via ultrasonication (\u003cb\u003eF\u003c/b\u003e) for [PSI+]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 prion fibrils. In \u003cb\u003eD\u003c/b\u003e, the fractions of Sup35 prion fibrils, which resisted the indicated temperatures, were determined by SDD-AGE. In \u003cb\u003eE\u003c/b\u003e, purified Sup35NM-GFP fibrils were incubated for the indicated time in 8 M urea, and the proportion of released monomeric Sup35NM-GFP was determined by SDS-PAGE, using a sample boiled without urea as a measure of the total Sup35NM-GFP amount in fibrils. In \u003cb\u003eF\u003c/b\u003e, equal aliquots of Sup35NM-GFP fibrils were ultrasonicated for different durations, and the change in fibril size distribution was assessed by SDD-AGE. In all cases, electrophoresis was followed by immunoblotting and densitometry of blots.\u003c/p\u003e \u003cp\u003e \u003cb\u003eG.\u003c/b\u003e Hsp104 binding to strong- and weak-type Sup35 prion aggregates assessed via their co-sedimentation from cell lysates. The relative quantity of Hsp104 and aggregated Sup35 in pellets was assessed by Western blotting, using [\u003cem\u003epsi\u003c/em\u003e-] cells as a negative control.\u003c/p\u003e \u003cp\u003eP-values were calculated using two-tailed Student\u0026rsquo;s t-test.\u003c/p\u003e \u003cp\u003e \u003cb\u003e2. The role of Core1A and Core1B in the maintenance of Sup35 fibrils stability\u003c/b\u003e \u003c/p\u003e \u003cp\u003eSince both strong and weak prion variants have the Core1 containing parts of similar length (residues 2\u0026ndash;72 overall), the observed difference in physical properties of strong- and weak-type amyloid fibrils \u003cem\u003ein vivo\u003c/em\u003e could not be explained by the different length of Core1. Paradoxically, the strong-type Sup35 fibrils, which are less stable than weak-type ones (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD-\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF), and more fragmentable \u003cem\u003ein vivo\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB), always have a more PK-resistant Core1B\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e than weak-type ones (which even can lack Core1B in rare cases). To explain this discrepancy, we hypothesize that in fibrils of strong prion variant Core1B plays an important role in stabilization of Core1A structure (which, itself, is inherently less stable), and appears to be more tightly associated with Core1A. In contrast, \u0026ldquo;weak\u0026rdquo; fibrils have a more stable fold of Core1A, and therefore Core1B is dispensable for the maintenance of their integrity (and could be more loosely attached to Core1A). To test this hypothesis, we, firstly, analyzed the effects on the [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prions of several uncharged-to-charged amino acid substitutions in the Core1A and Core1B areas, previously described as prion-destabilizing ones. To this end, we substituted the wild type \u003cem\u003eSUP35\u003c/em\u003e allele by the mutant ones in [\u003cem\u003ePSI\u003c/em\u003e+]-S7 or [\u003cem\u003ePSI\u003c/em\u003e+]-W2 cells by the plasmid shuffle.\u003c/p\u003e \u003cp\u003eThe G58D substitution, located in the Core1B area (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), was previously shown to cure only strong [\u003cem\u003ePSI\u003c/em\u003e+] variants\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Therefore, we anticipated that this mutation could destabilize either Core1B itself or its association with Core1A. Indeed, in our hands, \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eG58D\u003c/em\u003e\u003c/sup\u003e allele, even as a sole source of Sup35 protein in cells, did not completely cure the [\u003cem\u003ePSI\u003c/em\u003e+]-S7 prion, but significantly altered its properties: weakened its nonsense-suppressor phenotype (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB, S2A), increased proportion of soluble Sup35 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), and decreased mitotic stability (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH). The PK resistance of the Core1B was noticeably reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). These features may seem to indicate a shift in the [\u003cem\u003ePSI\u003c/em\u003e+]-S7\u003csup\u003eG58D\u003c/sup\u003e properties towards a weak prion variant. However, contrary to such interpretation, [\u003cem\u003ePSI\u003c/em\u003e+]-S7\u003csup\u003eG58D\u003c/sup\u003e polymers became less thermoresistant and more fragile compared to [\u003cem\u003ePSI\u003c/em\u003e+]-S7\u003csup\u003eWT\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF-\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG) - and, therefore, their manifestations \u003cem\u003ein vivo\u003c/em\u003e likely originated from the increased (rather than decreased, as in the case of \u0026ldquo;true\u0026rdquo; weak [\u003cem\u003ePSI\u003c/em\u003e+]) fragmentation by chaperones; Fig. S2H), leading to a partial amyloid disaggregation, as it was initially proposed by Pei et al\u003csup\u003e38\u003c/sup\u003e. These data support the hypothesis that the Core 1B region is important for the \u003cem\u003ein vivo\u003c/em\u003e stability of strong [\u003cem\u003ePSI\u003c/em\u003e+]. Reverse shuffle (Sup35\u003csup\u003eG58D\u003c/sup\u003e\u0026rarr; Sup35\u003csup\u003eWT\u003c/sup\u003e) led to restoration of the initial [\u003cem\u003ePSI\u003c/em\u003e+]-S7 phenotype (Fig. S2D-S2E), suggesting that Core 1B has regenerated its initial fold that is likely dictated by Core 1A.\u003c/p\u003e \u003cp\u003eIn contrast to the strong [\u003cem\u003ePSI\u003c/em\u003e+]-S7 prion, weak [\u003cem\u003ePSI\u003c/em\u003e+]-W2 variant was nearly unaffected by the Sup35\u003csup\u003eWT\u003c/sup\u003e\u0026rarr; Sup35\u003csup\u003eG58D\u003c/sup\u003e substitution, as follows from its PK resistance profile, nonsense suppressor phenotype, and physical properties of fibrils (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB-\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH, S2A). This suggests that Core1B is more loosely packed in the [\u003cem\u003ePSI\u003c/em\u003e+]-W2 fibrils, and, is more dispensable for structure stabilization and fibril\u0026rsquo;s physical properties. Earlier, we observed that a weak [\u003cem\u003ePSI\u003c/em\u003e+]-C3 isolate lacked protease resistance after residue Y45\u003csup\u003e39\u003c/sup\u003e. However, its thermal stability was even higher (Fig. S2F-S2G) than that of [\u003cem\u003ePSI\u003c/em\u003e+]-W2. This allows us to conclude that the region 46\u0026ndash;72, a major part of the Core1B, is unlikely to have a significant impact on the overall stability of weak-type fibrils.\u003c/p\u003e \u003cp\u003eAnti-prion substitutions S17R and Q24R\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e are located in the Core1A area (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The Sup35\u003csup\u003eWT\u003c/sup\u003e\u0026rarr; Sup35\u003csup\u003eS17R\u003c/sup\u003e shuffle cured both [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). The Sup35\u003csup\u003eWT\u003c/sup\u003e\u0026rarr; Sup35\u003csup\u003eQ24R\u003c/sup\u003e substitution also cured [\u003cem\u003ePSI\u003c/em\u003e+]-W2 variant, while [\u003cem\u003ePSI\u003c/em\u003e+]-S7 prion has acquired an unusual very weak nonsense-suppressor phenotype (dark-pink colony color, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB) with the absence of SDS-resistant polymers (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Since this phenotype did not emerge in the [\u003cem\u003epsi-\u003c/em\u003e] cells Sup35\u003csup\u003eWT\u003c/sup\u003e\u0026rarr; Sup35\u003csup\u003eQ24R\u003c/sup\u003e shuffle (Fig. S2B), and it was partially sensitive to Hsp104 inhibition (Fig. S2C), we suggest that it likely has prion nature, but the its prion structure is substantially altered and further study of its properties is beyond the scope of this article.\u003c/p\u003e \u003cp\u003eTherefore, we conclude that Core1A is of ultimate importance for maintenance of both strong and weak prions variants, while Core1B fold is dispensable for weak variants and important for strong variant as a stabilizer of strong-type prion polymers. Based on these conclusions, we suggest that Core1A itself has less robust and stable fold in the strong-type fibrils (compared to weak-type ones), and therefore, requires stabilizing interactions with Core1B.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eA.\u003c/b\u003e Schematic representation of the Sup35 N-domain showing its Q/N-rich and oligopeptide repeats (R1-R4) regions, location of the fully PK-resistant Core1A and partly PK-resistant Core1B, and the studied anti-prion mutations.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eB.\u003c/b\u003e Nonsense-suppressor and \u003cem\u003eAde+\u003c/em\u003e phenotypes of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells with the \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e gene shuffled for either \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eG58D\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ24R\u003c/em\u003e\u003c/sup\u003e or \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eS17R\u003c/em\u003e\u003c/sup\u003e mutant alleles.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eC.\u003c/b\u003e SDS-resistant Sup35 polymers detected by SDD-AGE in the lysates of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells where \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e was shuffled for the indicated mutant variants.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eD.\u003c/b\u003e Proportions of soluble Sup35 and Sup35\u003csup\u003eG58D\u003c/sup\u003e in [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells, estimated as by the Sup35 ability to enter a 10% polyacrylamide gel without boiling.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eE.\u003c/b\u003e Mapping of PK-resistant cores of prion fibrils isolated from [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells and composed of either Sup35NM\u003csup\u003eWT\u003c/sup\u003e-GFP or Sup35NM\u003csup\u003eG58D\u003c/sup\u003e-GFP proteins.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eF-G.\u003c/b\u003e Comparison of the resistance to heat (thermal stability assay, \u003cb\u003eF\u003c/b\u003e) and mechanical shearing by ultrasonication (\u003cb\u003eG\u003c/b\u003e) for strong- and weak-type prion polymers consisting of either Sup35NM\u003csup\u003eWT\u003c/sup\u003e-GFP or Sup35NM\u003csup\u003eG58D\u003c/sup\u003e-GFP proteins.\u003c/p\u003e \u003cp\u003e \u003cb\u003eH.\u003c/b\u003e Comparison of prion mitotic stability for [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells with the \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e allele replaced for the \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eG58D\u003c/em\u003e\u003c/sup\u003e or \u003cem\u003eSUP35\u003c/em\u003e\u003csup\u003e\u003cem\u003eQ24R\u003c/em\u003e\u003c/sup\u003e alleles.\u003c/p\u003e \u003cp\u003eP-values were calculated using two-tailed Student\u0026rsquo;s t-test\u003c/p\u003e \u003cp\u003e \u003cb\u003e3. Cryo-EM analysis of [\u003c/b\u003e \u003cb\u003ePSI+\u003c/b\u003e \u003cb\u003e]-S7 prion structure\u003c/b\u003e \u003c/p\u003e \u003cp\u003eNext, we sought to resolve the atomic structures of \u003cem\u003eex vivo\u003c/em\u003e Sup35NM-GFP fibrils purified from either [\u003cem\u003ePSI+\u003c/em\u003e]-S7 or [\u003cem\u003ePSI+\u003c/em\u003e]-W2 cells. For both variants, most of the fibrils consisted of a single protofilament of ~\u0026thinsp;6 nm width. For [\u003cem\u003ePSI+\u003c/em\u003e]-S7 fibrils, we observed alteration of nearly untwisted (~\u0026thinsp;54%) and clearly twisted (~\u0026thinsp;46%) segments on the same filaments (Fig. S4C). These segments displayed similar 2D class features apart from the degree of twist, indicating a common molecular fold. [\u003cem\u003ePSI+\u003c/em\u003e]-W2 fibrils had an almost untwisted morphology, which made them unsuitable for further structure reconstruction.\u003c/p\u003e \u003cp\u003eUsing the clearly twisted segments, we resolved the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 fibril core structure at 2.7 \u0026Aring; resolution (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH, S6A-S6D). The corresponding helical parameters were determined as a rise of 4.78 \u0026Aring; and a twist of \u0026minus;\u0026thinsp;1.39\u0026deg;, corresponding to a left-handed pitch of 123.5 nm. A single protomer exhibits a non-planar, undulating arrangement (with an axial span of ~\u0026thinsp;9.1 \u0026Aring; along the fibril axis) and forms an \u0026ldquo;amyloid key\u0026rdquo;-like fold\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e in the S2-N64 region. This span of the ordered structure is generally consistent with the length of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 Core1 as determined by PK mapping (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI). Residues 2\u0026ndash;4 precede the structured region and are partially stabilized. Among them, residue 4 is sufficiently constrained for its position to be assigned unambiguously, whereas residues 2 and 3 remain partially flexible, contributing diffuse density without discernible side-chain features.\u003c/p\u003e \u003cp\u003eThe protomer molecule is stabilized by numerous \u0026ldquo;horizontal\u0026rdquo; interactions (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE-\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF) between amino acid side chains, which stabilize the structure in the plane perpendicular to the fibril axis. An \u0026ldquo;inner\u0026rdquo; zipper, formed by a short β-arch spanning Y13\u0026ndash;G25, stabilizes the central Q/N-rich subdomain of the core (Core1A). An \u0026ldquo;outer\u0026rdquo; zipper is generated by the antiparallel packing of N8\u0026ndash;Q18 against Q47\u0026ndash;Y63, thereby linking Core1A to the peripheral subdomain (Core1B) and integrating both into a single ordered fold. Because the protomer is non-planar, these interfaces are volumetric rather than flat: side chains interdigitate with a slight axial offset, producing three-dimensional packing that reduces solvent access and increases shape complementarity (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). In both zippers, the antiparallel β-strands are staggered by approximately half of the helical rise, a feature typical of amyloid steric zippers that promotes volumetric packing. Quantitative analysis of interface geometry shows that the inner and outer steric zipper interfaces both display high shape complementarity (SC)\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e values of 0.81\u0026ndash;0.82, in contrast to the lower value (0.76) displayed by the interfaces located outside of the zipper-forming regions (Fig. S6G). Notably, the SC values observed for the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 steric zippers approach those reported for the microcrystals formed by the classical Sup35-derived peptide GNNQQNY (SC\u0026thinsp;=\u0026thinsp;0.86), where the highly complementary polar side chains form completely dry steric zipper interfaces\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. The order of the corresponding sidechains packing is also similar between the GNNQQNY microcrystal and the outer zipper of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 amyloid (Fig. S6H).\u003c/p\u003e \u003cp\u003eAlong the fibril axis, stabilization is provided by both canonical in-register cross-β hydrogen-bonding network between backbone amides and carbonyls, as well as by some \u0026ldquo;vertical\u0026rdquo; interactions of the sidechains: twelve tyrosine residues form intermolecular \u0026#120529;-columns, while numerous asparagine and glutamine residues form amide ladders (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). A more detailed description of fibril-stabilizing interactions and some additional structural features is provided in the \u003cem\u003eSupplementary Results\u003c/em\u003e section. Notably, the major PK-resistant peptides from [\u003cem\u003ePSI+\u003c/em\u003e]-S7 digestion start from the N-terminus (S2 residue) and end at the positions Y32, Y35, Q38, A42 and Y45\u003csup\u003e22\u003c/sup\u003e, which belong to the region between the \u0026ldquo;inner\u0026rdquo; and \u0026ldquo;outer\u0026rdquo; steric zippers (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH). As the core structure predominantly consists of polar residues (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD), its surface is largely hydrophilic and uncharged (Fig. S6E-S6F).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eA.\u003c/b\u003e Cross-sectional view of the 3D reconstruction of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid core structure.\u003c/p\u003e \u003cp\u003e \u003cb\u003eB-C.\u003c/b\u003e Rendered 3D model of the ordered part of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid fibril (\u003cb\u003eB\u003c/b\u003e), and the atomic model of a single protomer superimposed on the sharpened cryo-EM density map (\u003cb\u003eC\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eD.\u003c/b\u003e The topology of the [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid fold, showing the location of β-strands. The sidechains are color-coded based on their physiochemical properties.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eE.\u003c/b\u003e Atomic model of the [\u003cem\u003ePSI\u003c/em\u003e+]-S7 amyloid core showing the \u0026ldquo;inner\u0026rdquo; and \u0026ldquo;outer\u0026rdquo; steric zipper elements.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eF.\u003c/b\u003e Close-up of the parts of \u0026ldquo;inner\u0026rdquo; and \u0026ldquo;outer\u0026rdquo; steric zippers showing their \u0026ldquo;staggered\u0026rdquo; organization along the Z-axis of the fibril.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eG.\u003c/b\u003e Close-up of several \u0026#120529;-columns formed by tyrosine residues (green arrows) and amide ladders formed by glutamine residues (violet arrows) stabilizing the core structure along the axis of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 prion fibrils.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eH-I.\u003c/b\u003e Comparison of the PK-protected region and cryo-EM-solved ordered structure of Core1. Dotted lines indicate the main stop sites for proteinase K digestion of Core1B. The \u0026ldquo;inner\u0026rdquo; and \u0026ldquo;outer\u0026rdquo; steric zipper zones are highlighted in magenta and blue, respectively. The major PK-resistant peptides observed in the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 preparation are shown below.\u003c/p\u003e \u003cp\u003e \u003cb\u003e4. Thermodynamic profiling of [\u003c/b\u003e \u003cb\u003ePSI+\u003c/b\u003e \u003cb\u003e]-S7 atomic structure and its interaction with the anti-prion point mutations in Sup35 N-domain.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo make a structure-based assessment of the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 fibrils stability, we estimated their free energy of unfolding (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e) using the FoldX algorithm\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. For comparison, we also calculated the solvation energy (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e), which has been used to estimate the amyloid structures stability in many published works\u003csup\u003e\u003cspan additionalcitationids=\"CR45 CR46 CR47\" citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. To get a broader picture, we also calculated \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e and \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e for a set of published amyloid structures belonging to different partially overlapping biological and physicochemical groups. We found that [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structure has \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e comparable to that of some structures of known functional amyloids, and higher (i.e. less favorable) than that of pathology-related amyloids (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). In contrast, its \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e is very high (i.e. non-favorable) even in comparison with the majority of functional amyloids. Notably, the only known amyloid structure with even higher \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e compared with Sup35 fibrils is the Orb2 functional amyloid of \u003cem\u003eD. melanogaster\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. Given that both of these structures are dominated by stabilizing steric zipper motifs (as in the case with many other highly stable pathological amyloid structures), we suggest that the \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e parameter is much less accurate for assessing amyloid structure stability than the \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e parameter, and seems to greatly underestimate the true stability of amyloid structures consisting mostly of polar residues (Fig. S7).\u003c/p\u003e \u003cp\u003eWe further used the resolved [\u003cem\u003ePSI+\u003c/em\u003e]-S7 amyloid core structure to provide structural rationale for the effect of known anti-prion mutations in Sup35 N-terminal domain. To this end, we used a set of 15 such mutations that mostly cause uncharged-to-charged residue changes\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. A majority of these substitutions are located in the areas of either \u0026ldquo;inner\u0026rdquo; or \u0026ldquo;outer\u0026rdquo; steric zippers (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), suggesting that the zipper regions are most important for the structure stability. Using FoldX, we estimated the fibril stability change upon these mutations, and found all substitutions to be significantly destabilizing (i.e., increasing \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). Remarkably, by testing another 10 manually selected uncharged-to-charged residue substitutions located in non-zipper-forming regions, we found their effect on [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structure stabilization energy to be negligible. Thus, [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structural analysis indicates which structural motifs are most sensitive to substitutions. One could also suggest that introducing charged residues into the tightly locked zippers would loosen their packing, thereby decreasing fibril stability (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eA.\u003c/b\u003e Gibbs free energy of unfolding and solvation energy of Sup35/[\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion structure (red arrow) in comparison with that of several representatives of different amyloid classes.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eB.\u003c/b\u003e Amyloid core structure with known anti-prion mutations\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. Most of them represent uncharged-to-charged residues substitutions in the zipper-forming regions.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eC.\u003c/b\u003e Predicted change in the free energy of unfolding (ΔΔG) calculated by FoldX software for the anti-prion Sup35 mutations in comparison with the wild type structure. In contrast to known anti-prion mutations, manually selected substitutions (uncharged-to-charged residues) located outside of the zipper-forming regions have only a minor impact on prion stability. P-value was calculated using two-tailed Student\u0026rsquo;s t-test.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e\u003cp\u003e5. Comparison of ex vivo Sup35 prion structure with in vitro formed Sup35 amyloids.\u003c/p\u003e\u003cp\u003eThe work of Tanaka et al. published recently\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, describes the cryo-EM structures of Sc4 and Sc37 amyloids that are presumed to be structural equivalents of the strong and weak Sup35 prions, respectively. Comparison of the structures of Sc4 amyloid with that of the strong [\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion reveals both a significant similarity and a substantial difference (Fig.\u0026nbsp;5A-5B). The similarity relates to a region 2\u0026ndash;26, which we consider to be the most significant and defining part of the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structure. The highest similarity is observed in the area of Y13\u0026ndash;G25 β-arch (Fig.\u0026nbsp;5B, S6G). A ꞵ-strand N27-Q30 is also found in both structures, but it is oriented differently. In contrast to [\u003cem\u003ePSI+\u003c/em\u003e]-S7, in Sc4 structure the major part of the N36-N64 stretch (roughly corresponding to the Core1B area) is fully or partially disordered. The exception is the Q47-S53 stretch, which forms short ꞵ-strand attached to the top of Y13-G25 hairpin by the zipper-like interactions. However, the registry of these interactions in [\u003cem\u003ePSI+\u003c/em\u003e]-S7 and Sc4 structures is different.\u003c/p\u003e \u003cp\u003eSurprisingly, the structure of Sc37 amyloid also bears noticeable similarity to that of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion: both structures contain long zipper-like element between the antiparallel stretches Q6-Q23 and Y45-N63 (Fig.\u0026nbsp;5A, 5C, S6G), and similar overall span of the Core1 (residues 5\u0026ndash;65 and 2\u0026ndash;64, respectively). The thermodynamic parameters (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e and \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e) are also similar between [\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion and Sc37 amyloid structures, as well as between Sc4 and the Core1A part of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structure (Fig.\u0026nbsp;5D). Tanaka et al. also reported that about a half of the Sc37 amyloid segments also contained the additional core (K102-Q132); this subtype of Sc37 structure has \u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e ~25% lower compared with that of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structure.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003e. \u003cb\u003eComparison of\u003c/b\u003e \u003cb\u003eex vivo\u003c/b\u003e \u003cb\u003eSup35 prion structure with\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e \u003cb\u003eformed Sup35 amyloids\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003eA.\u003c/b\u003e Structure of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 \u003cem\u003eex vivo\u003c/em\u003e prion fibrils in comparison with the Sc4 (PDB ID: 9XBN) and Sc37 (PDB ID: 9XBO) amyloids formed \u003cem\u003ein vitro\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Steric zipper elements are highlighted in grey.\u003c/p\u003e \u003cp\u003e \u003cb\u003eB-C.\u003c/b\u003e Structural alignment of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 with Sc4 (\u003cb\u003eB\u003c/b\u003e) and Sc37 (\u003cb\u003eC\u003c/b\u003e) structures, with the close-up of some regions fully or partially similar between [\u003cem\u003ePSI+\u003c/em\u003e]-S7 with Sc4 structures.\u003c/p\u003e \u003cp\u003e \u003cb\u003eD.\u003c/b\u003e Comparison of the free energy of unfolding (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eUnf\u003c/sub\u003e) and solvation energy (\u003cem\u003eΔG\u003c/em\u003e\u003csub\u003eSolv\u003c/sub\u003e) between [\u003cem\u003ePSI+\u003c/em\u003e]-S7, Sc4 and Sc37 structures.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eThe mechanism of prion fragmentation\u003c/h2\u003e \u003cp\u003ePreviously, two hypotheses were proposed explaining what defines the difference in the frequency of fragmentation between the weak and strong Sup35 prions. One hypothesis proposed that the more frequent fragmentation of the strong prion is due to its lower mechanical strength. The other assumed that since Hsp104 is a powerful molecular machine with 12 ATPase units, the limiting factor is not its ability to unfold a prion, but its ability to find its target, which occurs with the help of Hsp70 (Ssa1/2) and Hsp40 (Sis1). In support of the latter model, it was observed that progressive deletions of the oligopeptide repeats in the Sup35 N domain, which were assumed to be chaperone targets, reduced prion fragmentation, thus increasing prion size and eventually caused prion loss\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e,\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e. However, the set of data, presented in this and several other works\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e, supports the mechanical strength being the key factor. This also agrees with the recent finding that the Sup35 interaction with chaperones mainly occurs through the Ssa1/2 binding site located beyond the prion-forming region, at Sup35 residues 143-164\u003csup\u003e10\u003c/sup\u003e. While one can note that the mechanical strength model was generally more popular than its alternative, it was based predominantly on the assumption of the equivalence of synthetic Sup35 amyloids and \u003cem\u003ein vivo\u003c/em\u003e Sup35 prions, which proves to be incorrect according to this and some previous studies.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCryo-EM structure of Sup35 prion\u003c/h3\u003e\n\u003cp\u003eMany mammalian amyloid fibrils are composed of two or more protofilaments. In contrast, the Sup35 polymers extracted from both [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 cells consisted of a single filament. This rules out the scenario where the higher stability of weak [PSI+] filaments is related to their double- or triple-protofilament arrangement. \u003cem\u003eIn vitro\u003c/em\u003e, Sup35 usually forms single filaments, though twin filaments were also observed\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. However, it appears likely that only single-filament fibrils could succeed as viable prions in yeasts. Propagation of yeast prions relies on the ability of Hsp104 to fragment prion fibers\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. While in single-filament fibers this requires the extraction of just one protomer, in multi-protofilament fibers several protomers opposing each other should be extracted.\u003c/p\u003e \u003cp\u003eHere we found that [\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion has the parallel-in-register architecture. This confirms the earlier conclusion of Wickner and coauthors\u003csup\u003e\u003cspan additionalcitationids=\"CR57\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e, and disproves the alternative solenoid-type model\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e,\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. Of note, the solenoid model was formulated basing on study of \u003cem\u003ein vitro\u003c/em\u003e generated Sup35 fibrils\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e, for which the parallel in-register architecture was also shown recently\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Importantly, PK digestion patterns of strong and weak prions are similar within these groups\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, suggesting that many (if not all) \u003cem\u003ein vivo\u003c/em\u003e [\u003cem\u003ePSI+\u003c/em\u003e] variants are likely to share similar strong or weak fibril architecture.\u003c/p\u003e \u003cp\u003eIn this work we showed that the Core1A (residues 2\u0026ndash;32) is the key element of the strong and weak prion structures. While some known anti-prion substitutions are located in the Core1B region\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e, genetic screen performed by DePace et al yielded the anti-prion substitutions located mostly between residues 8 and 24\u003csup\u003e40\u003c/sup\u003e, which corresponds well with the location of inner and outer steric zippers (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Our data (Fig. S2D) and a previous report\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e suggest that the Core1A region affects the folding of the rest of the structure. While the dominance of the Core1A appears to be largely related to its physical strength, our recent study shows that the terminal location itself is a significant factor helping any sequence to influence folding of the remaining part of prion structure. The most telling observation was that a random 23 residue sequence added to the Sup35 N-terminus altered the fold and overall span of the Sup35 prion structures in their original location\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cb\u003eRelation between the\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e \u003cb\u003eformed Sup35 amyloids and the prion propagating\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eEarlier works revealed a significant difference of the PK maps of the strong and weak Sup35 prions with their presumed counterparts generated \u003cem\u003ein vitro\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. This work, together with Tanaka et al\u003csup\u003e32\u003c/sup\u003e resolves this discrepancy by establishing cryo-EM structures and showing that the PK maps correspond well to these structures, while the structures differ significantly.\u003c/p\u003e \u003cp\u003eThe Sc4 and [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structures are very similar in their most important fully PK-resistant part, or Core1A, but have little in common for the rest of the structures. Therefore, when yeast is transfected with Sc4 amyloid, its region 32\u0026ndash;64 has to rearrange, acquiring the continuous cross-β fold (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Alternatively, a conformer similar to the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 prion may be present in Sc4 preparations as a minor fraction, becoming the most abundant in yeast (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). A recent study using \u003cem\u003ein vivo\u003c/em\u003e NMR indicates that both of these processes occur with \u003cem\u003ein vitro\u003c/em\u003e-formed α-synuclein amyloid: upon transfection into cells, the proportion of major and minor isoforms reversed, and the disordered flanking regions of the fibrils were remodelled\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur observations with the Sup35 G58D anti-prion suggest why the [\u003cem\u003ePSI+\u003c/em\u003e]-S7 fold may have selective preference in yeast cells. This substitution destabilizes the steric zipper which attaches Core1B to the Core1A, making the whole prion structure less stable (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE-\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB-\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), which should increase prion disaggregation and promote prion loss\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Apparently, the Sc4 structure, which almost lacks this steric zipper, should be even less stable than the strong prion with G58D substitution and thus unable to propagate in yeast.\u003c/p\u003e \u003cp\u003eAlthough the cryo-EM structure of the weak [\u003cem\u003ePSI+\u003c/em\u003e] variant has not been solved, some important notes about the relation of the Sc37 structure and the natural weak [\u003cem\u003ePSI+\u003c/em\u003e] prions can be made. Comparison of the Sc37 and [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structures explains why their PK digestion patterns are so similar. Both structures include similar, though not identical steric zippers formed by stretches N8-Y16 and Q50-Y63, which ensures high resistance to PK of the region Y45-Q72 in both cases. However, in all studied weak [\u003cem\u003ePSI+\u003c/em\u003e] prions, and [\u003cem\u003ePSI+\u003c/em\u003e]-W2 in particular, this region shows low PK resistance\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eSumming up, in contrast to the Sc4 case, it is difficult to see an easy way for the Sc37 structure to rearrange into a weak prion structure. It was shown that Sc37 transfection into yeast produces a weak [\u003cem\u003ePSI+\u003c/em\u003e] variant similar to the VK variant\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, which exhibits a PK digestion pattern typical for all weak variants\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, but different from that of Sc37\u003csup\u003e30\u003c/sup\u003e. Thus, it appears likely that Sc37 preparation contains a rare polymorph, structurally distinct from the published Sc37 major structure, that can be preferentially selected in cells and give rise to a weak [\u003cem\u003ePSI+\u003c/em\u003e] prion.\u003c/p\u003e \u003cp\u003e \u003cb\u003eRelation between the length of Sup35 amyloid core structures and PK-resistant peptides.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eMapping of amyloid cores through partial proteinase K digestion is an affordable supplement and in some senses an alternative to the full-atomic structure reconstruction. Now, PK maps are available for several \u003cem\u003eex vivo\u003c/em\u003e yeast prions\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e,\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e, but their full-atomic structures are missing. Thus, it is of interest to understand how PK maps relate to actual structures. This work and those of Tanaka group\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e provide a rare opportunity to learn this.\u003c/p\u003e \u003cp\u003eThe ordered core of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 fibrils defined by the cryo-EM (residues 2\u0026ndash;64) is smaller than the PK-resistant region (residues 2\u0026ndash;72). Similar differences were observed for the Sc37 amyloid (5\u0026ndash;65 versus 2\u0026ndash;72)\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Sc4 has an ordered core comprising residues 4\u0026ndash;35 and 47\u0026ndash;52, while the PK-resistant region includes residues 2\u0026ndash;46; cryo-EM structure of S17R4 amyloid includes residues 88\u0026ndash;129 versus PK-resistant core at residues 81\u0026ndash;148; S17R37 cryo-EM structure includes residues 69\u0026ndash;132 and PK-resistant core at residues 62\u0026ndash;144. Thus, PK-protected regions are about 7\u0026ndash;12 residues longer at each end excluding the N-terminus, which has no such length of unstructured sequence. The existence of such a gap could be explained by the fact that the PK active site is located within a pit of the PK molecule\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e, and to be digested at its end, an unfolded region of certain length needs to enter this site and to reach the catalytic center. Still, it is unclear how PK can cut within the structured region at residues 32 to 64, and this mechanism requires further investigation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eA-B.\u003c/b\u003e Two possible explanations of relation between the structures of \u003cem\u003ein vitro\u003c/em\u003e formed Sc4 amyloid and \u003cem\u003ein vivo\u003c/em\u003e [\u003cem\u003ePSI+\u003c/em\u003e]-strong prion variant. First one (\u003cb\u003eA\u003c/b\u003e) implies structural rearrangement of Sc4 amyloid by the extension of existing cross-β fold on some unfolded parts of Sup35 molecule (the N36-Y46 and G54-N64 stretches), and the second one (\u003cb\u003eB\u003c/b\u003e) implies positive selection of rare, but the most fitted Sc4 polymorphs inside the yeast cells.\u003c/p\u003e \u003cp\u003e \u003cb\u003eC.\u003c/b\u003e Proposed mechanism of the anti-prion action of the G58D and Q24R substitutions with respect to the strong [\u003cem\u003ePSI+\u003c/em\u003e] variant structure resolved in this work.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eYeast strains and cultivation conditions\u003c/h2\u003e \u003cp\u003eThe work used derivatives of the 74-D694 yeast strain (\u003cem\u003eMATa, ade1-14, trp1-289, his3Δ-200, ura3-52, leu2-3,112\u003c/em\u003e) carrying either [\u003cem\u003ePSI\u003c/em\u003e+]-S7 or [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prion. To prevent the \u003cem\u003ede novo\u003c/em\u003e formation of any alternative amyloid conformers in the experiments with Sup35NM-GFP overproduction, the \u003cem\u003eRNQ1\u003c/em\u003e gene in these strains was disrupted by \u003cem\u003eHIS3\u003c/em\u003e insertion. For the plasmid shuffle experiments, derivatives of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 strains were created, where the chromosomal \u003cem\u003eSUP35\u003c/em\u003e gene was disrupted by \u003cem\u003eTRP1\u003c/em\u003e insertion and the \u003cem\u003eSUP35\u003c/em\u003e gene introduced on the centromeric pRS316-SUP35 vector.\u003c/p\u003e \u003cp\u003eThe standard yeast media were used. Synthetic complete media (SCM) contained 6.7 g/L yeast nitrogen base, 20 g/L glucose or 24 g/L galactose, and required amino acids. For colony color development, SCM contained reduced amount of adenine (7 mg/L, or 1/3 of standard). For overproduction of proteins of interest under control of the \u003cem\u003eGAL1\u003c/em\u003e promoter, we used SCM-Gal medium with 24 g/L galactose instead of glucose and 100 mg/L of adenine. Rich YPD medium contained 10 g/L yeast extract, 20 g/L peptone, 20 g/L glucose and 20 g/L agar when required. For better development of the colony color phenotype, solid YPD-red medium was used (6 g/L yeast extract (Oxoid), 20 g/L peptone, 20 g/L glucose and 20 g/L agar). Cells were grown at 30\u0026deg;C.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eYeast genetics assays\u003c/h2\u003e \u003cp\u003eYeast transformation with DNA was performed using the standard LiAc-PEG method\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. Yeast transfection with prion fibrils (\u0026ldquo;spheroplast transformation\u0026rdquo;) was performed as described by Tanaka et al\u003csup\u003e27\u003c/sup\u003e, using an \u0026ldquo;empty\u0026rdquo; pRS316 vector as a selection marker.\u003c/p\u003e \u003cp\u003eTo switch the Sup35\u003csup\u003eWT\u003c/sup\u003e allele to Sup35\u003csup\u003eMUT\u003c/sup\u003e, we used the plasmid shuffle technology. Cells where the genomic \u003cem\u003eSUP35\u003c/em\u003e was deleted, and the episomal Sup35\u003csup\u003eWT\u003c/sup\u003e was encoded on the centromeric pRS316 (\u003cem\u003eURA3\u003c/em\u003e) plasmid, were transformed with pRS315 (\u003cem\u003eLEU2\u003c/em\u003e) plasmid encoding the Sup35\u003csup\u003eMUT\u003c/sup\u003e allele. These \u003cem\u003eSUP35\u003c/em\u003e genes included the native promoter. Transformants were grown on selective medium for 3 days, and then streaked on the complete synthetic medium containing 1 mg/ml of fluoroorotic acid (FOA) to ensure loss of the \u003cem\u003eURA3\u003c/em\u003e plasmid. After 3-day growth, the cells from fully-grown colonies were pooled, streaked to single cells on the YPD-red medium, grown for 3 days, and the loss of plasmid encoding the Sup35\u003csup\u003eWT\u003c/sup\u003e allele was confirmed by inability to grow on the synthetic medium lacking uracil.\u003c/p\u003e \u003cp\u003eTo make yeast spots, cells were grown overnight in liquid YPD, then cell suspensions were diluted to OD600\u0026thinsp;=\u0026thinsp;1, and 8 \u0026micro;l of suspensions were spotted to the YPD-red plates. For better development of colony color, spots were grown at 30\u0026deg;C for 3 days and then incubated at 4\u0026deg;C for 1 day prior imaging.\u003c/p\u003e \u003cp\u003eThe propagon number in [\u003cem\u003ePSI\u003c/em\u003e+] cells was evaluated as described by Cox et al\u003csup\u003e34\u003c/sup\u003e. Briefly, cells were streaked on YPD plates, grown at 30\u0026deg;C for 3 days, freshly-grown colonies were streaked on YPD plates with 4 mM GuHCl and grown for 1.5-2 days. Small colonies were collected and streaked on the complete synthetic medium containing 0.4 mg/ml adenine (2% of standard concentration). At this adenine concentration, only the cells that inherited [\u003cem\u003ePSI\u003c/em\u003e+] prion can form fully-grown colonies. Plates were incubated at 30\u0026deg;C for 5 days, and the number of fully-grown colonies corresponding to each colony from the YPD-GuHCl plate was calculated. It is assumed that this value corresponds to the number of propagons (the elementary units of prion inheritance) in a single yeast cell carrying [\u003cem\u003ePSI\u003c/em\u003e+] prion.\u003c/p\u003e \u003cp\u003eTo estimate the level of prion stability upon the mitotic cell growth, several [\u003cem\u003ePSI\u003c/em\u003e+] colonies were pooled, streaked to single cells on YPD plate and grown at 30\u0026deg;C for 3 days. Then 5 individual colonies (as individual biological replicates) were separately streaked to single cells on YPD-red plated, grown at 30\u0026deg;C for 3 days, then incubated 1 day at 4\u0026deg;C, and the proportion of red (i.e. [\u003cem\u003epsi\u003c/em\u003e-]) colonies were determined for each replicate. As the single cell takes\u0026thinsp;~\u0026thinsp;25 generations to form a fully-grown colony, the obtained values correspond to the percentage of prion loss after 25 generations of mitotic growth.\u003c/p\u003e \u003cp\u003eTo demonstrate the increased level of the weak [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prion loss upon the short-term heat shock, cells carrying [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prions were grown in liquid YPD medium to the log phase (OD600\u0026thinsp;~\u0026thinsp;1.5), then incubated at 45\u0026deg;C for 30 min, streaked to the single cells on YPD-red plates and grown at 30\u0026deg;C for 3 days. The proportions of red and red/pink sectored colonies were calculated.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePurification of Sup35NM-GFP prion fibrils from yeast for cryo-EM analysis.\u003c/b\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMethod consideration and creating the protein construct\u003c/h3\u003e\n\u003cp\u003eIn [\u003cem\u003ePSI\u003c/em\u003e+] cells with the native production level of endogenous Sup35 protein, prions exist as a population of short fibrils\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e, which are not suitable for the cryo-EM structure reconstruction. To overcome this issue and facilitate prion purification we overproduced the GFP-tagged construct containing prionogenic part of Sup35 protein (residues 1-239) using the multicopy plasmid carrying the \u003cem\u003eSUP35NM-GFP\u003c/em\u003e gene under the control of \u003cem\u003eGAL1\u003c/em\u003e promoter. Upon this procedure, this construct adopts the pre-existing prion fold\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e and form long filaments sequestered into the large higher-order aggregate\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn this and previous work\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e we observed that Sup35NM-GFP fibrils \u003cem\u003eper se\u003c/em\u003e are poorly suitable for cryo-EM reconstruction due to the structural noise imposed by GFP globules and the disordered Sup35 M domain (a. a. 125\u0026ndash;239). These parts were removed by trypsin. However, the obtained fibrils were prone to precipitation, so that only a small proportion of them were separate on cryo-EM grids and suitable for structure reconstruction. To cause repulsion of fibrils, we introduced an acidic cluster DDDNEDSEEDDEDGGP\u003cb\u003eR\u003c/b\u003eGSR between G96 and A155 residues, resulting in the Sup35DE-GFP protein (Figure S4A). Using MALDI, we confirmed that trypsin cleaves this protein at arginine residue (in bold), but does not cleave at the only upstream R28 located inside the prion core.\u003c/p\u003e\n\u003ch3\u003ePurification of Sup35DE-GFP fibrils\u003c/h3\u003e\n\u003cp\u003eYeast cells carrying either [\u003cem\u003ePSI+\u003c/em\u003e]-S7 or [\u003cem\u003ePSI+\u003c/em\u003e]-W2 prions were transformed with two plasmids: the pYES2-SUP35DE-GFP plasmid for overproduction of prion protein, and the rescue plasmid pRS315-SUP35C which reduced the cellular toxicity associated with Sup35 overproduction in the [\u003cem\u003ePSI\u003c/em\u003e+] cells, thereby increasing the level of Sup35DE-GFP production. Cells were grown to the stationary phase in selective SCM-glucose medium, then a two-fold volume of selective SCM-Gal medium was added, cells were incubated for 15 hours and harvested. Cell pellets (2*4 ml in 50 ml tubes) were mixed with an equal volume of glass beads and a half volume of lysis buffer (TBS (Tris Buffered Saline) buffer, 5 mM PMSF, 1 mM DTT). Cells were broken by vortexing at maximum speed for 10 minutes at 4\u0026ordm;C. Cell lysates were transferred to 15-ml tubes and spun down at 2500g for 15 min at 4\u0026ordm;C. The supernatant lacked GFP and was discarded, and the upper layer of pellet containing GFP was resuspended in 4 ml of fresh TBS buffer supplemented with 100 \u0026micro;g/ml RNaseA, 50 \u0026micro;g/ml DNase I, placed on top of sucrose gradient #1 (TBS based, 0.5 ml of 60% sucrose and 3 ml of 30% sucrose in 15 ml Falcon tube), and spun for 15 min at 2500g, 4\u0026ordm;C. The GFP-containing fraction of gradient was collected and sucrose was diluted two-fold to reduce sucrose concentration.\u003c/p\u003e \u003cp\u003eTo disassemble higher-order clumps of Sup35DE-GFP prion polymers into individual fibrils, SDS was added to the final concentration of 4%. Samples were incubated for 20 min at 10\u0026ordm;C to avoid both protein degradation and SDS precipitation. The dissolution of visible Sup35DE-GFP clumps into individual fibrils was monitored with the fluorescent microscope Axioskop 40 (Zeiss, Germany). Then, samples were centrifuged for 1 min at 21300 g to remove the remaining debris, and GFP-containing supernatant was placed on top of sucrose gradient #2 (based on TBS buffer with 0.3% Sarcosyl: 100 \u0026micro;l of 70% sucrose\u0026thinsp;+\u0026thinsp;200 \u0026micro;l of 60% sucrose\u0026thinsp;+\u0026thinsp;700 \u0026micro;l of 30% sucrose in 3.5 ml tube) and spun down in Beckman Coulter Optima Max ultracentrifuge SW50 rotor at 260 000 g for 4 h at 10\u0026ordm;C. GFP-tagged prion fibrils were trapped in the 60% sucrose fraction. The GFP-positive fraction of the gradient was collected, frozen in liquid nitrogen and stored at -70\u0026ordm;C.\u003c/p\u003e \u003cp\u003eTo remove GFP and \u0026ldquo;fuzzy coat\u0026rdquo; part of Sup35DE-GFP fibrils, the samples were diluted two-fold with TBS buffer and treated with 0.1 mg/ml trypsin (Sigma) for 60 min at 37\u0026deg;C. To terminate reaction, 1 mM PMSF was added after 1 h incubation. To remove the unwanted digestion products, the preparations were centrifuged at 260000 g for 4 h at 4\u0026deg;C using sucrose gradient #2. Pure fibrils were found in the 60% sucrose fraction. This fraction was dialyzed against 1 L of 0.1X TBS buffer with 0.1% Sarcosyl. The samples were then concentrated\u0026thinsp;~\u0026thinsp;10-fold by evaporation of water excess for 1 h at room temperature. The scheme of fibril isolation, trypsin treatment and purification is shown in Figures S4A and S4B. The purity of the final sample, as determined by the SDS-PAGE, was ~\u0026thinsp;95% (Fig. S3C). The Sup35DE fibrils showed reasonably small aggregation on the cryo-EM grids, suitable for cryo-EM data collection.\u003c/p\u003e \u003cp\u003eTo ensure that prion strain properties were not changed upon Sup35DE-GFP overproduction and protein purification procedure, we transfected the 74-D694 [\u003cem\u003epsi\u003c/em\u003e-] cells with either [\u003cem\u003ePSI\u003c/em\u003e+]-S7 or [\u003cem\u003ePSI\u003c/em\u003e+]-W2 fibrils, extracted and treated as described above. In both cases, initial prion phenotype (either strong or weak nonsense suppression) was reproduced (Fig. S4B).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCryo-electron microscopy\u003c/h2\u003e \u003cp\u003e \u003cem\u003eSample preparation and data acquisition\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eQuantifoil R1.2/1.3 Cu grids were glow-discharged for 20 s at 20 mA (0.26 mbar) using a PELCO easiGlow system. 3 \u0026micro;l of the fibril suspension (0.38 mg/ml) were applied to the grids, blotted for 3 s at 100% humidity and 4\u0026deg;C, and plunge-frozen in liquid ethane using a Vitrobot Mark IV (Thermo Fisher Scientific, USA). The grid was then stored in liquid nitrogen until use.\u003c/p\u003e \u003cp\u003eCryo-EM data were collected on a Titan Krios 60\u0026ndash;300 transmission electron microscope (Thermo Fisher Scientific, USA) equipped with a field emission electron gun X-FEG (Thermo Fisher Scientific, USA), spherical-aberration corrector (CEOS GmbH, Germany), a post-column BioQuantum energy filter (Gatan, USA) and a K3 direct electron detector (Gatan, USA) in counting mode using SerialEM 4.055 and 9-hole (2 exposures per hole) image-shift data acquisition strategy at the National Research Centre \u0026lsquo;\u0026lsquo;Kurchatov Institute\u0026rsquo;\u0026rsquo;. The microscope was operated at 300 kV with a nominal magnification of 81000x, corresponding to a pixel size of 0.863 \u0026Aring; at the specimen level, and an electron energy selecting slit of 20 eV. A total dose of 52 e⁻/\u0026Aring;\u0026sup2; within a 3.5 s exposure time was fractionated into 70 frames, resulting in an electron dose of 0.74 e⁻/\u0026Aring;\u0026sup2; per frame. A total of 9,015 movies were collected in a nominal defocus range from \u0026minus;\u0026thinsp;0.7 to -2.0 \u0026micro;m with a step of 0.1 \u0026micro;m. Detailed parameters of data acquisition are listed in Table S1.\u003c/p\u003e \u003cp\u003e \u003cem\u003eImage processing.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eMovies were motion-corrected and CTF parameters estimated in Warp. Micrograph power spectra showed a characteristic\u0026thinsp;~\u0026thinsp;4.8 \u0026Aring; meridional reflection, indicative of well-ordered cross-β structure. Filaments were automatically picked with crYOLO using an inter-box step of 24 \u0026Aring; (five helical units per step).\u003c/p\u003e \u003cp\u003eHelical segments were extracted with box sizes of 1024 and 512 pixels (binned to 256 and 128 for the initial steps) to support accurate crossover measurements and efficient 2D classification, respectively. 1024-pixel and 512-pixel particles were used for initial 2D classification, helical pitch estimation, and ab initio model generation. The 2D classes from 1024-pixel boxes reported a crossover distance of ~\u0026thinsp;70 nm. After initial 2D classification with 512-pixel boxes, false positives and broken filaments were removed and the dataset was partitioned into twisted (585,256 segments) and low-twist (674,012 segments) subsets. For homogeneous processing, only the twisted subset was retained. An initial 3D model of the twisted fibrils was then generated using \u003cem\u003erelion_helix_inimodel2d\u003c/em\u003e, assuming a left-handed helix, and used as input for subsequent refinement.\u003c/p\u003e \u003cp\u003eStarting from 585,256 segments extracted with a 384-pixel box, an initial helical refinement in CryoSPARC yielded a reconstruction at 3.6 \u0026Aring;, which improved to 3.23 \u0026Aring; after local refinement with a soft mask around the fibril core. This was followed by two consecutive rounds of 3D classification without alignment in CryoSPARC, each combined with cycles of CTF refinement, helical refinement, and local refinement. In the first round, a well-defined subset comprising 163,538 segments produced a 2.81 \u0026Aring; reconstruction (helical parameters: rise 4.78 \u0026Aring;, twist \u0026minus;\u0026thinsp;1.278\u0026deg;). A second round of classification of this subset was performed to further purge residual heterogeneity, yielding 50,658 high-quality segments. Refinement of this final subset in CryoSPARC resulted in a global resolution of 2.69 \u0026Aring;, as determined by the gold-standard FSC 0.143 criterion. The best-defined map corresponded to fibrils exhibiting a rise of 4.782 \u0026Aring; and a twist of \u0026minus;\u0026thinsp;1.388\u0026deg;. The map was sharpened with a negative B-factor of \u0026minus;\u0026thinsp;58 \u0026Aring;\u0026sup2;.\u003c/p\u003e \u003cp\u003e \u003cem\u003eModel Building and Refinement.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eA single protomer was built \u003cem\u003ede novo\u003c/em\u003e into the sharpened map in Coot, with unambiguous sequence assignment from residue 4 to residue 64. Density preceding residue 4 was poorly resolved: residues 2\u0026ndash;3 were therefore modeled according to sequence context (initiator Met1 absent; Ser2 Nα-acetylated; Asp3 present), while acknowledging their flexibility in the map. The built protomer was stacked into a five-layer segment by rigid-body placement in UCSF Chimera, and subsequently subjected to PHENIX real-space refinement with NCS constraints between the five chains and individual ADPs. Model-to-map and stereochemical validation statistics are summarized in Table S1. Shape complementarity (SC) values were calculated using the \u003cem\u003esc\u003c/em\u003e utility from the CCP4 suite for selected pairs of residue segments forming lateral β-sheet interfaces, with molecular surfaces constructed from five neighboring fibril layers to reduce edge effects.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003ePK resistance analysis of Sup35NM-GFP prion fibrils\u003c/h2\u003e \u003cp\u003eYeast cells carrying either [\u003cem\u003ePSI+\u003c/em\u003e]-S7 or [\u003cem\u003ePSI+\u003c/em\u003e]-W2 prions were transformed with pYES2-SUP35NM-GFP plasmid (where Sup35NM sequence was either WT or containing G58D substitution) and grown to the stationary phase in the selective SCM-glucose medium. Then a two-fold volume of selective SCM-Gal medium was added, cells were incubated for 15 hours and harvested.\u003c/p\u003e \u003cp\u003eTo map the PK-resistant regions of Sup35NM-GFP fibrils, we extracted and purified the Sarcosyl-resistant fractions from large pellets of yeast cells (~\u0026thinsp;3\u0026ndash;5 grams per preparation), as described in the original work\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Briefly, yeast cells were disrupted with glass beads, fraction of Sup35NM-GFP aggregates was isolated, mixed with Sarkosyl (final concentration\u0026thinsp;=\u0026thinsp;5%), sonicated, loaded onto a 20\u0026ndash;65% sucrose gradient, and ultracentrifuged at 260000g for 4 hours at 4\u0026ordm;C. The GFP-containing fraction was collected, and the resulting Sup35NM-GFP sample (diluted to a final concentration of 200 \u0026micro;g/ml) was digested with PK (25 \u0026micro;g/ml) for 1 hour at RT. The PK-resistant part of the fibrils was then precipitated with 40% acetone, resuspended in water, boiled, and analyzed by MALDI-TOF/TOF mass spectrometer UltrafleXtreme (Bruker, Germany). PK has no sequence specificity except it does not cut before and after proline\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. Peptides were identified by tandem mass spectrometry (MS-MS) and/or as groups of related peaks. As we have previously shown\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, the N-terminal methionine of Sup35PD-GFP was completely removed and replaced by an acetyl group in the yeast cells, and thus all N-terminal peptides started from amino acid residue № 2. Raw MALDI data were analyzed using Bruker flexAnalysis 3.3 software, and PrK resistance was calculated and plotted as a function (Rn) of the normalized PrK resistance index for a given sequence position (n), given by:\u003c/p\u003e \u003cp\u003e \u003cb\u003eRn = \u0026sum;\u003c/b\u003e \u003cb\u003eS\u003c/b\u003e\u003csub\u003e\u003cb\u003en\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e/ \u0026sum;\u003c/b\u003e \u003cb\u003eS\u003c/b\u003e\u003csub\u003e\u003cb\u003emax\u003c/b\u003e\u003c/sub\u003e\u003c/p\u003e \u003cp\u003ewhere the \u003cb\u003e\u0026sum;\u003c/b\u003e \u003cb\u003eS\u003c/b\u003e\u003csub\u003e\u003cb\u003en\u003c/b\u003e\u003c/sub\u003e is a sum of the areas of all MS peaks containing this amino acid position (n), and \u003cb\u003e\u0026sum;\u003c/b\u003e \u003cb\u003eS\u003c/b\u003e\u003csub\u003e\u003cb\u003emax\u003c/b\u003e\u003c/sub\u003e \u0026ndash; the maximum peak area sum value.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eSDD-AGE analysis of amyloid polymers\u003c/h2\u003e \u003cp\u003eYeast cells were grown in 20 ml of the YPD media to the mid-log phase, collected to 2.4 ml Eppendorf tubes and lysed by vigorous vortexing with glass beads and 100 \u0026micro;l of TBS buffer supplemented with 5 mM PMSF, 1 mM DTT and Complete protease inhibitor cocktail (Roche) at 4\u0026deg;C for 10 min. Cell lysates were mixed with 4X SDD-AGE sample buffer (1X is 0.5X Tris acetate/EDTA, 2% SDS, 5% glycerol, and 0.05% Bromophenol blue) and subjected to the agarose gel electrophoresis, as described by Kryndushkin et al\u003csup\u003e33\u003c/sup\u003e. Protein samples resolved in agarose gel were then vacuum-transferred to a nitrocellulose membrane (Porablot, MACHEREY-NAGEL), and subjected to a standard Western-Blotting procedure using anti-Sup35NM rabbit polyclonal primary antibodies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eThermal stability assay\u003c/h2\u003e \u003cp\u003eThe procedure was performed as standard SDD-AGE, but with some modifications\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Cells were grown and lysed as described above. To avoid possible effect of cell lysate components on the thermal stability, the Sup35 aggregates were purified by centrifugation at 20000 rpm for 30 min at 4\u0026deg;C. The supernatant was discarded, and the pellet was dissolved at room temperature in 300 \u0026micro;l of SDD-AGE sample buffer. Remaining cell debris was spun down at 4000 rpm for 1 min, and 50 \u0026micro;l sample aliquots were transferred to PCR tubes and incubated at either 25, 50, 60, 70, 85, or 99\u0026deg;C for 8 minutes, then cooled to 10\u0026deg;C and loaded onto a 1.8% agarose gel (gel strength\u0026thinsp;=\u0026thinsp;1000 g/sm\u003csup\u003e2\u003c/sup\u003e) with 0.1% SDS and analyzed by SDD-AGE followed by Western blotting with anti-Sup35NM primary polyclonal antibodies. The integrated intensity of Sup35NM-GFP smears, as a measure of the amount of Sup35NM-GFP amyloids that withstood heating, was then quantified using Fiji software. For each amyloid preparation, the experiment was performed in 3\u0026ndash;4 replicates, and mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM values were calculated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eMechanical fragmentation of Sup35NM-GFP fibrils\u003c/h2\u003e \u003cp\u003eThe fibrils of Sup35NM-GFP (with either WT sequence of Sup35NM, or containing G58D substitution) were extracted from the yeast cells using the protocol similar to that used for fibril purification for the cryo-EM study: the big clumps of long fibrils were isolated from the yeast lysates, dissolved into single fibrils with 4% SDS, and the fraction of purified fibrils was then obtained by ultracentrifugation in sucrose gradient. The obtained fractions were dialyzed overnight against the TBS buffer, obtained samples were diluted 20-fold and divided into 0.5 ml aliquotes. One aliquot was leaved unsonicated, while others were sonicated (2/2 second on/off cycle, 50% amplitude) using VibraCell sonicator (Sonics \u0026amp; Material Inc, USA) for indicated times at +\u0026thinsp;4\u0026deg;C. All samples were then mixed with SDD-AGE sample buffer and subjected to 1.8% agarose gel electrophoresis, followed by Western blotting and immunostaining with anti-Sup35NM primary antibodies. The distributions of Sup35NM-GFP signal on the gel, which reflect the Sup35NM-GFP fibril mobility and the shift in their size upon sonication, were quantified using densitometry analysis with FIJI software.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eUrea denaturation assay\u003c/h2\u003e \u003cp\u003eSup35NM-GFP fibrils, templated \u003cem\u003ein vivo\u003c/em\u003e by either [\u003cem\u003ePSI\u003c/em\u003e+]-S7 or [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prion variant, were isolated as described in the \u0026ldquo;PK resistance analysis of Sup35NM-GFP amyloid fibrils\u0026rdquo; chapter. The final samples were diluted 20-fold in TAE buffer with a designated urea concentration, and incubated for 8 h at 37\u0026deg;C without agitation. After that, the Laemmli sample buffer was added and samples were subjected to a standard SDS-PAGE procedure followed by Western blotting with immunostaining against Sup35NM. At this stage, only the monomeric Sup35NM-GFP molecules, that were released from the polymers upon their urea-mediated denaturation, are capable of entering the polyacrylamide gel. As a measure of total Sup35NM-GFP amount in polymers, similar samples without urea were boiled for 5 min, leading to a total conversion of polymers into the monomers. We used a set of urea concentrations (i.e., 0M, 4M, 6M and 8M) and found that both [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 polymers started to dissolve only in 8M urea solution, and the dissolution was much more prominent in the case of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 polymers (Fig. S1C). Further, we made a time-lapse experiment, where the aliquots of either [\u003cem\u003ePSI\u003c/em\u003e+]-S7- or [\u003cem\u003ePSI\u003c/em\u003e+]-W2-templated fibrils were incubated at 37\u0026deg;C for 3\u0026ndash;72 hours, and then analyzed as described above. The proportion of polymers dissolved in urea was calculated as a ratio of the amount of monomeric protein released in 8 M urea samples, to the amount of monomeric protein released in the boiled sample w/o urea. The experiment was performed in triplicates, and mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM values were calculated for both types of fibrils at each time point.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eChaperones binding assay\u003c/h2\u003e \u003cp\u003eFor this experiment, we used isogenic cells with either [\u003cem\u003ePSI+\u003c/em\u003e]-S7, [\u003cem\u003ePSI+\u003c/em\u003e]-W2 or [\u003cem\u003epsi-\u003c/em\u003e] prion status, grown to a logarithmic phase. Cells were lysed using a standard lysis procedure with glass beads in 100 \u0026micro;l of TAE buffer supplemented with 10 mM PMSF, 1M NaN\u003csub\u003e3\u003c/sub\u003e and 0.01 mg/mL RNAseA. NaN\u003csub\u003e3\u003c/sub\u003e and RNAseA were added to disrupt large polyribosome complexes which can include monomeric Sup35 in the [\u003cem\u003epsi\u003c/em\u003e-] cells and, thus, to prevent monomeric Sup35 sedimentation upon the centrifugation on the later stages. Lysates were transferred into separate tubes, incubated for 10 min at 15\u0026deg;C, diluted to 1000 \u0026micro;l by TAE buffer, and spun at 6000 g for 2 min to remove unlysed cells and cell debris. 600 \u0026micro;l of supernatants were transferred into separate tubes and centrifuged at 20000 g for 15 min, 4\u0026deg;C, to precipitate prion aggregates with bound Hsp104. Supernatants were removed, 50 \u0026micro;l of 1X SDS-PAGE loading buffer was added, and the pellets were resuspended by vigorous pipetting. Samples were then boiled for 5 min (to dissolve Sup35 amyloid fibrils into monomers), cooled, centrifuged for 1 min at 20000 g, and then 5 \u0026micro;l aliquots were subjected to a standard SDS-PAGE procedure followed by immunoblotting. Each sample was immunostained by the primary polyclonal antibodies against either Sup35NM or Hsp104. The relative amounts of both Sup35 and Hsp104 proteins in each sample (in the arbitrary units) were determined by the densitometry of protein bands, and the relative Hsp104 binding to Sup35 prion aggregates was assessed in arbitrary units as:\u003c/p\u003e \u003cp\u003e \u003cb\u003eB\u003c/b\u003e \u003csup\u003e \u003cb\u003eX\u003c/b\u003e \u003c/sup\u003e \u003cb\u003e= (N\u003c/b\u003e\u003csup\u003e\u003cb\u003eX\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003ePSI+\u003c/b\u003e\u003cb\u003e]\u0026minus;V\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Hsp104) - N\u003c/b\u003e\u003csup\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003epsi\u0026minus;\u003c/b\u003e\u003cb\u003e]\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Hsp104))/(N\u003c/b\u003e\u003csup\u003e\u003cb\u003eX\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003ePSI+\u003c/b\u003e\u003cb\u003e]\u0026minus;V\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Sup35) - N\u003c/b\u003e\u003csup\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003epsi\u0026minus;\u003c/b\u003e\u003cb\u003e]\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Sup35))\u003c/b\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cb\u003eB\u003c/b\u003e\u003csup\u003e\u003cb\u003eX\u003c/b\u003e\u003c/sup\u003e is a relative Hsp104 binding for a given replicate of given [\u003cem\u003ePSI+\u003c/em\u003e] variant (\u0026ldquo;\u003cem\u003e[PSI+]-V\u003c/em\u003e\u0026rdquo;), \u003cb\u003eN\u003c/b\u003e\u003csup\u003e\u003cb\u003eX\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003ePSI\u003c/b\u003e\u003cb\u003e+]\u0026minus;V\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Hsp104)\u003c/b\u003e is a Hsp104 band intensity for a given replicate of given [\u003cem\u003ePSI+\u003c/em\u003e] variant, \u003cb\u003eN\u003c/b\u003e\u003csup\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003epsi\u003c/b\u003e\u003cb\u003e\u0026minus;]\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Hsp104)\u003c/b\u003e is a mean Hsp104 band intensity for [\u003cem\u003epsi-\u003c/em\u003e] cells, \u003cb\u003eN\u003c/b\u003e\u003csup\u003e\u003cb\u003eX\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[\u003c/b\u003e\u003cb\u003ePSI\u003c/b\u003e\u003cb\u003e+]\u0026minus;V\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Sup35)\u003c/b\u003e is a Sup35 band intensity for a given replicate of given [\u003cem\u003ePSI+\u003c/em\u003e] variant, and \u003cb\u003eN\u003c/b\u003e\u003csup\u003e\u003cb\u003eMean\u003c/b\u003e\u003c/sup\u003e\u003csub\u003e\u003cb\u003e[psi\u0026minus;]\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e(Sup35)\u003c/b\u003e is a mean Sup35 band intensity for [\u003cem\u003epsi-\u003c/em\u003e] cells.\u003c/p\u003e \u003cp\u003eThe experiment was performed in four biological replicates for each cell type, and mean\u0026thinsp;+\u0026thinsp;SEM values were calculated for [\u003cem\u003ePSI\u003c/em\u003e+]-S7 and [\u003cem\u003ePSI\u003c/em\u003e+]-W2 prion variants. Importantly, only barely detectable Hsp104 amounts were found in the pellet of [\u003cem\u003epsi\u003c/em\u003e-] cells, indicating that almost all Hsp104 molecules detected in the pellets of [\u003cem\u003ePSI\u003c/em\u003e+] cells sedimented due to the binding to Sup35 prion aggregates.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eStructure-based energy calculations\u003c/h2\u003e \u003cp\u003eTo calculate the free energy of unfolding (ΔG\u003csub\u003eUnf\u003c/sub\u003e), and stability change upon mutations (i.e., the difference in free energy of unfolding between the fibrils formed by the mutated and wild-type Sup35 protein), we used FoldX 5.1 plugin in the YASARA program.\u003c/p\u003e \u003cp\u003eΔG\u003csub\u003eUnf\u003c/sub\u003e was calculated (default temperature\u0026thinsp;=\u0026thinsp;298 K, default ionic strength\u0026thinsp;=\u0026thinsp;0.05 M) using the equation:\u003c/p\u003e \u003cp\u003e \u003cb\u003eΔG\u003c/b\u003e \u003csub\u003e \u003cb\u003eUnf\u003c/b\u003e \u003c/sub\u003e \u003cb\u003e= W\u003c/b\u003e\u003csub\u003e\u003cb\u003evdw\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e\u0026sdot;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003evdw\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e+ W\u003c/b\u003e\u003csub\u003e\u003cb\u003esolvH\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e\u0026sdot;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003esolvH\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e+ W\u003c/b\u003e\u003csub\u003e\u003cb\u003esolvP\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e\u0026sdot;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003esolvP\u003c/b\u003e\u003c/sub\u003e\u0026thinsp;\u003cb\u003e+\u0026thinsp;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003ewb\u003c/b\u003e\u003c/sub\u003e\u0026thinsp;\u003cb\u003e+\u0026thinsp;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003ehbond\u003c/b\u003e\u003c/sub\u003e\u0026thinsp;\u003cb\u003e+\u0026thinsp;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003eel\u003c/b\u003e\u003c/sub\u003e\u0026thinsp;\u003cb\u003e+\u0026thinsp;ΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003eKon\u003c/b\u003e\u003c/sub\u003e \u003cb\u003e+ W\u003c/b\u003e\u003csub\u003e\u003cb\u003emc\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e\u0026sdot;T\u0026sdot;ΔS\u003c/b\u003e\u003csub\u003e\u003cb\u003emc\u003c/b\u003e\u003c/sub\u003e\u0026thinsp;\u003cb\u003e+\u0026thinsp;+\u0026thinsp;W\u003c/b\u003e\u003csub\u003e\u003cb\u003esc\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e\u0026sdot;T\u0026sdot;ΔS\u003c/b\u003e\u003csub\u003e\u003cb\u003esc\u003c/b\u003e\u003c/sub\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cb\u003eΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003evdw\u003c/b\u003e\u003c/sub\u003e is the sum of the van der Waals contributions of all atoms with respect to the same interactions with the solvent; \u003cb\u003eΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003esolvH\u003c/b\u003e\u003c/sub\u003e and \u003cb\u003eΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003esolvP\u003c/b\u003e\u003c/sub\u003e are the differences in solvation energy for apolar and polar groups respectively when these change from the unfolded to the folded state; \u003cb\u003eΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003ehbond\u003c/b\u003e\u003c/sub\u003e is the free energy difference between the formation of an intra-molecular hydrogen bond compared to inter-molecular hydrogen-bond formation (with solvent); \u003cb\u003eΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003ewb\u003c/b\u003e\u003c/sub\u003e is the extra stabilising free energy provided by a water molecule making more than one hydrogen bond to the protein (water bridges) that cannot be taken into account with non-explicit solvent approximations; \u003cb\u003eΔG\u003c/b\u003e\u003csub\u003e\u003cb\u003eel\u003c/b\u003e\u003c/sub\u003e is the electrostatic contribution of charged groups, including the helix dipole; \u003cb\u003eΔS\u003c/b\u003e\u003csub\u003e\u003cb\u003emc\u003c/b\u003e\u003c/sub\u003e is the entropy cost of fixing the backbone in the folded state; this term is dependent on the intrinsic tendency of a particular amino acid to adopt certain dihedral angles; \u003cb\u003eΔS\u003c/b\u003e\u003csub\u003e\u003cb\u003esc\u003c/b\u003e\u003c/sub\u003e the entropic cost of fixing a side chain in a particular conformation.\u003c/p\u003e \u003cp\u003eFor a comparison with the Sup35 amyloid core structure solved in this work, the ΔG\u003csub\u003eUnf\u003c/sub\u003e values were also calculated for a set of manually selected amyloid structures representing different, partially overlapping physicochemical and biological groups: prions, pathological amyloids, functional amyloids, amyloid structures enriched in polar residues, and amyloid structures enriched in hydrophobic residues. To obtain more consistent results, all types of fibrils were standardized by the length (to be 5-mers), and the calculated ΔG\u003csub\u003eUnf\u003c/sub\u003e values were normalized for a single layer of the fibril.\u003c/p\u003e \u003cp\u003eThe solvation energy (ΔG\u003csub\u003eSolv\u003c/sub\u003e) values were calculated using the Amyloid Illustrator package\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e (default pH\u0026thinsp;=\u0026thinsp;7). As in case of ΔG\u003csub\u003eUnf\u003c/sub\u003e, the solvation energy was normalized for a single layer and compared with the values calculated for a set of published amyloid structures representing different biological and physicochemical groups.\u003c/p\u003e \u003cp\u003eTo calculate the change in ΔG\u003csub\u003eUnf\u003c/sub\u003e upon mutations (stability change, ΔΔG), we used the model of [\u003cem\u003ePSI\u003c/em\u003e+]-S7 fibril comprising 5 protomers, and introduced the corresponding residue substitution in all of these protomers (using the \u0026ldquo;Mutate multiple residues\u0026rdquo; feature in FoldX5.1; T\u0026thinsp;=\u0026thinsp;298K, ionic strength\u0026thinsp;=\u0026thinsp;0.05M, pH\u0026thinsp;=\u0026thinsp;7, VdW design\u0026thinsp;=\u0026thinsp;2). ΔΔG was determined as a difference between the ΔG\u003csub\u003eUnf\u003c/sub\u003e values of mutated and WT molecules. The resulting values were normalized for a single layer of the fibril. The positive values correspond to decreased structure stability, and negative values correspond to increased structure stability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eImage representation\u003c/h2\u003e \u003cp\u003eImage representations of reconstructed densities and atomic models were created with UCSF Chimera\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e, ChimeraX\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e and the Amyloid Illustrator package\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. To make an alignment between [\u003cem\u003ePSI\u003c/em\u003e+]-S7 (this work), Sc4 (PDB ID: 9XBN) and Sc37 (PDB ID: 9XBO) structures, we used the Pairwise Structure Alignment tool in the \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ewww.rcsb.org\u003c/span\u003e\u003c/span\u003e\u003cspan address=\"http://www.rcsb.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e web-site (TM-align algorithm).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eData availability\u003c/h2\u003e \u003cp\u003eAtomic coordinates of [\u003cem\u003ePSI+\u003c/em\u003e]-S7 structure have been deposited in the Protein Data Bank under accession number 21BQ. The corresponding cryo-EM density map has been deposited in the Electron Microscopy Data Bank under accession code EMD-67558.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Resource Center of Probe and Electron Microscopy of National Research Center \u0026ldquo;Kurchatov Institute\u0026rdquo; (http://rc.nrcki.ru/pages/main/nanozond/facilities/12604/index.shtml). This work has been carried out using computing resources of the federal collective usage center Complex for Simulation and Data Processing for Mega-science Facilities at National Research Center \u0026ldquo;Kurchatov Institute\u0026rdquo; (http://ckp.nrcki.ru). MALDI mass spectrometry and DNA sequencing were performed by the Shared Access Equipment Centre \u0026ldquo;Industrial Biotechnology\u0026rdquo; of the FRC \u0026ldquo;Fundamentals of Biotechnology\u0026rdquo; (RAS). We thank Valery Urakov for the spheroplast prion transformation and Andrey Moiseenko (Biology Faculty of Moscow State University, Moscow) for the help with negative staining electron microscopy experiments.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by the Russian Science Foundation, grant #23-74-00062 (A.A.D, A.D.B, V.V.K), and in part by the Ministry of Science and Higher Education of the Russian Federation (O.M.V, K.M.B, V.O.P). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eConceptualization: A.A.D, V.V.K; methodology: A.A.D, Y.M.C., A.D.B., V.V.K, K.M.B; acquisition of data: A.A.D, Y.M.C., A.D.B., O.V.M, T.N.B.; analysis of data: Y.M.C., A.A.D, A.D.B, V.V.K, T.N.B.; writing \u0026mdash; original draft: A.A.D; writing \u0026mdash; review and editing: V.V.K, A.A.D, Y.M.C., A.D.B., K.M.B., V.O.P; visualization: A.A.D, Y.M.C.; supervision: V.V.K, A.A.D, K.M.B; funding acquisition: V.V.K, V.O.P. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eStefani M, Dobson CM (2003) Protein aggregation and aggregate toxicity: new insights into protein folding, misfolding diseases and biological evolution. J Mol Med 81:678\u0026ndash;699\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEisenberg D, Jucker M (2012) The Amyloid State of Proteins in Human Diseases. Cell 148:1188\u0026ndash;1203\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSergeeva AV, Galkin AP (2020) Functional amyloids of eukaryotes: criteria, classification, and biological significance. Curr Genet 66:849\u0026ndash;866\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOtzen D, Riek R (2019) Functional Amyloids. Cold Spring Harb Perspect Biol 11:a033860\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalkin AP, Sysoev EI, Valina AA (2023) Amyloids and prions in the light of evolution. Curr Genet 69:189\u0026ndash;202\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickner RB, Masison DC, Edskes HK (1995) [PSI] and [URE3] as yeast prions. Yeast 11:1671\u0026ndash;1685\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaushkin SV, Kushnirov VV, Smirnov VN, Ter-Avanesyan MD (1996) Propagation of the yeast prion-like [psi+] determinant is mediated by oligomerization of the SUP35-encoded polypeptide chain release factor. EMBO J 15:3127\u0026ndash;3134\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatino MM, Liu JJ, Glover JR, Lindquist S (1996) Support for the prion hypothesis for inheritance of a phenotypic trait in yeast. Science 273:622\u0026ndash;626\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiebman SW, Chernoff YO (2012) Prions Yeast Genetics 191:1041\u0026ndash;1072\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen CH et al (2024) Exposed Hsp70-binding site impacts yeast Sup35 prion disaggregation and propagation. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 121, e2318162121\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollinge J, Clarke AR (2007) A General Model of Prion Strains and Their Pathogenicity. Science 318:930\u0026ndash;936\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollinge J (2016) Mammalian prions and their wider relevance in neurodegenerative diseases. Nature 539:217\u0026ndash;226\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJucker M, Walker LC (2013) Self-propagation of pathogenic protein aggregates in neurodegenerative diseases. Nature 501:45\u0026ndash;51\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoedert M (2015) Alzheimer\u0026rsquo;s and Parkinson\u0026rsquo;s diseases: The prion concept in relation to assembled Aβ, tau, and α-synuclein. Science 349:1255555\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoedert M, Masuda-Suzukake M, Falcon B (2017) Like prions: the propagation of aggregated tau and α-synuclein in neurodegeneration. Brain 140:266\u0026ndash;278\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiang W, Yau W-M, Lu J-X, Collinge J, Tycko R (2017) Structural variation in amyloid-β fibrils from Alzheimer\u0026rsquo;s disease clinical subtypes. Nature 541:217\u0026ndash;221\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSawaya MR, Hughes MP, Rodriguez JA, Riek R, Eisenberg DS (2021) The expanding amyloid family: Structure, stability, function, and pathogenesis. Cell 184:4857\u0026ndash;4873\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheres SHW, Ryskeldi-Falcon B, Goedert M (2023) Molecular pathology of neurodegenerative diseases by cryo-EM of amyloids. Nature 621:701\u0026ndash;710\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhurana V, Lindquist S (2010) Modelling neurodegeneration in Saccharomyces cerevisiae: why cook with baker\u0026rsquo;s yeast? Nat Rev Neurosci 11:436\u0026ndash;449\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKochneva-Pervukhova NV et al (2001) [PSI+] prion generation in yeast: characterization of the strain difference. Yeast 18:489\u0026ndash;497\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUptain SM, Sawicki GJ, Caughey B, Lindquist S (2001) Strains of [PSI(+)] are distinguished by their efficiencies of prion-mediated conformational conversion. EMBO J 20:6236\u0026ndash;6245\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDergalev AA, Alexandrov AI, Ivannikov RI, Ter-Avanesyan MD, Kushnirov VV (2019) Yeast Sup35 Prion Structure: Two Types, Four Parts, Many Variants. Int J Mol Sci 20\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang Y-W, King C-Y (2019) A complete catalog of wild-type Sup35 prion variants and their protein-only propagation. Curr Genet. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00294-019-01003-8\u003c/span\u003e\u003cspan address=\"10.1007/s00294-019-01003-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChernoff Y, Lindquist S, Ono B, Inge-Vechtomov S, Liebman S (1995) Role of the chaperone protein Hsp104 in propagation of the yeast prion-like factor [psi+]. Science 268:880\u0026ndash;884\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTipton KA, Verges KJ, Weissman JS (2008) In vivo monitoring of the prion replication cycle reveals a critical role for Sis1 in delivering substrates to Hsp104. Mol Cell 32:584\u0026ndash;591\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShorter J, Lindquist S (2008) Hsp104, Hsp70 and Hsp40 interplay regulates formation, growth and elimination of Sup35 prions. EMBO J 27:2712\u0026ndash;2724\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTanaka M, Chien P, Naber N, Cooke R, Weissman JS (2004) Conformational variations in an infectious protein determine prion strain differences. Nature 428:323\u0026ndash;328\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTanaka M, Collins SR, Toyama BH, Weissman JS (2006) The physical basis of how prion conformations determine strain phenotypes. Nature 442:585\u0026ndash;589\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eToyama BH, Kelly MJS, Gross JD, Weissman JS (2007) The structural basis of yeast prion strain variants. Nature 449:233\u0026ndash;237\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhhashi Y et al (2018) Molecular basis for diversification of yeast prion strain conformation. Proc Natl Acad Sci U S A 115:2389\u0026ndash;2394\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlexandrov AI, Polyanskaya AB, Serpionov GV, Ter-Avanesyan MD, Kushnirov VV (2012) The effects of amino acid composition of glutamine-rich domains on amyloid formation and fragmentation. PLoS ONE 7:e46458\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTanaka M et al (2025) How Sup35 monomer conformation and amyloid fibril polymorphism determine yeast strain phenotypes. Preprint at. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21203/rs.3.rs-7945345/v1\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-7945345/v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKryndushkin DS, Alexandrov IM, Ter-Avanesyan MD, Kushnirov VV (2003) Yeast [ \u003cem\u003ePSI\u003c/em\u003e \u003csup\u003e+\u003c/sup\u003e ] Prion Aggregates Are Formed by Small Sup35 Polymers Fragmented by Hsp104. J Biol Chem 278:49636\u0026ndash;49643\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCox B, Ness F, Tuite M (2003) Analysis of the generation and segregation of propagons: entities that propagate the [PSI+] prion in yeast. Genetics 165:23\u0026ndash;33\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBagriantsev SN, Gracheva EO, Richmond JE, Liebman SW (2008) Variant-specific [PSI+] infection is transmitted by Sup35 polymers within [PSI+] aggregates with heterogeneous protein composition. Mol Biol Cell 19:2433\u0026ndash;2443\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDerkatch IL, Bradley ME, Zhou P, Liebman SW (1999) The PNM2 mutation in the prion protein domain of SUP35 has distinct effects on different variants of the [PSI+] prion in yeast. Curr Genet 35:59\u0026ndash;67\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerges KJ, Smith MH, Toyama BH, Weissman JS (2011) Strain conformation, primary structure and the propagation of the yeast prion [PSI+]. Nat Struct Mol Biol 18:493\u0026ndash;499\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePei F, DiSalvo S, Sindi SS, Serio TR (2017) A dominant-negative mutant inhibits multiple prion variants through a common mechanism. PLoS Genet 13:e1007085\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDergalev AA, Urakov VN, Agaphonov MO, Alexandrov AI, Kushnirov VV (2021) Dangerous Stops: Nonsense Mutations Can Dramatically Increase Frequency of Prion Conversion. IJMS 22:1542\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDePace AH, Santoso A, Hillner P, Weissman JS (1998) A Critical Role for Amino-Terminal Glutamine/Asparagine Repeats in the Formation and Propagation of a Yeast Prion. Cell 93:1241\u0026ndash;1252\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLawrence MC, Colman PM (1993) Shape Complementarity at Protein/Protein Interfaces. J Mol Biol 234:946\u0026ndash;950\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSawaya MR et al (2007) Atomic structures of amyloid cross-beta spines reveal varied steric zippers. Nature 447:453\u0026ndash;457\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchymkowitz J et al (2005) The FoldX web server: an online force field. Nucleic Acids Res 33:W382\u0026ndash;W388\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu J et al (2024) Cryo-EM structures of the D290V mutant of the hnRNPA2 low-complexity domain suggests how D290V affects phase separation and aggregation. J Biol Chem 300:105531\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao Q, Boyer DR, Sawaya MR, Ge P, Eisenberg DS (2019) Cryo-EM structures of four polymorphic TDP-43 amyloid cores. Nat Struct Mol Biol 26:619\u0026ndash;627\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosenberg GM et al (2023) Fibril structures of TFG protein mutants validate the identification of TFG as a disease-related amyloid protein by the IMPAcT method. PNAS Nexus 2:pgad402\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Ib\u0026aacute;\u0026ntilde;ez A et al (2022) Molecular interactions of FG nucleoporin repeats at high resolution. Nat Chem 14:1278\u0026ndash;1285\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEisenberg D, McLachlan AD (1986) Solvation energy in protein folding and binding. Nature 319:199\u0026ndash;203\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHervas R et al (2020) Cryo-EM structure of a neuronal functional amyloid implicated in memory persistence in \u003cem\u003eDrosophila\u003c/em\u003e. Science 367:1230\u0026ndash;1234\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarchante R, Rowe M, Zenthon J, Howard MJ, Tuite MF (2013) Structural Definition Is Important for the Propagation of the Yeast [PSI+] Prion. Mol Cell 50:675\u0026ndash;685\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKing C-Y (2001) Supporting the structural basis of prion strains: induction and identification of [PSI] variants. J Mol Biol 307:1247\u0026ndash;1260\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParham SN (2001) Oligopeptide repeats in the yeast protein Sup35p stabilize intermolecular prion interactions. EMBO J 20:2111\u0026ndash;2119\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShkundina IS, Kushnirov VV, Tuite MF, Ter-Avanesyan MD (2006) The Role of the N-Terminal Oligopeptide Repeats of the Yeast Sup35 Prion Protein in Propagation and Transmission of Prion Variants. Genetics 172:827\u0026ndash;835\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang Y, Kushnirov VV, King C (2021) Mutable yeast prion variants are stabilized by a defective Hsp104 chaperone. Mol Microbiol 115:774\u0026ndash;788\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKushnirov VV, Ter-Avanesyan MD (1998) Structure and Replication of Yeast Prions. Cell 94:13\u0026ndash;16\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShewmaker F, Wickner RB, Tycko R (2006) Amyloid of the prion domain of Sup35p has an in-register parallel beta-sheet structure. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 103, 19754\u0026ndash;19759\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShewmaker F, Kryndushkin D, Chen B, Tycko R, Wickner RB (2009) Two Prion Variants of Sup35p Have In-Register Parallel β-Sheet Structures, Independent of Hydration. Biochemistry 48:5074\u0026ndash;5082\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGorkovskiy A, Thurber KR, Tycko R, Wickner RB (2014) Locating folds of the in-register parallel β-sheet of the Sup35p prion domain infectious amyloid. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 111, E4615-4622\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrishnan R, Lindquist SL (2005) Structural insights into a yeast prion illuminate nucleation and strain diversity. Nature 435:765\u0026ndash;772\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoberts BE et al (2009) A synergistic small-molecule combination directly eradicates diverse prion strain structures. Nat Chem Biol 5:936\u0026ndash;946\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKing C-Y, Diaz-Avalos R (2004) Protein-only transmission of three yeast prion strains. Nature 428:319\u0026ndash;323\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalliamov AA, Urakov VN, Dergalev AA, Kushnirov VV (2025) On the Significance of the Terminal Location of Prion-Forming Regions of Yeast Proteins. \u003cem\u003eIJMS\u003c/em\u003e 26, 1637\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnsari S et al (2024) In cell NMR reveals cells selectively amplify and structurally remodel amyloid fibrils. Preprint at. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/2024.09.09.612142\u003c/span\u003e\u003cspan address=\"10.1101/2024.09.09.612142\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalliamov AA, Malukhina AD, Kushnirov VV (2024) Mapping of Prion Structures in the Yeast Rnq1. \u003cem\u003eIJMS\u003c/em\u003e 25, 3397\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBetzel C et al (2001) Structure of a Serine Protease Proteinase K from \u003cem\u003eTritirachium album limber\u003c/em\u003e at 0.98 \u0026Aring; Resolution. Biochemistry 40:3080\u0026ndash;3088\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGietz RD, Schiestl RH (2007) High-efficiency yeast transformation using the LiAc/SS carrier DNA/PEG method. Nat Protoc 2:31\u0026ndash;34\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaibil HR et al (2012) Heritable yeast prions have a highly organized three-dimensional architecture with interfiber structures. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 109, 14906\u0026ndash;14911\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurtseva AD et al (2023) Electron Microscopy Study of the Structure of the Sup35 Prion from Yeast Saccharomyces cerevisiae. Crystallogr Rep 68:872\u0026ndash;878\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePettersen EF et al (2004) UCSF Chimera\u0026mdash;A visualization system for exploratory research and analysis. J Comput Chem 25:1605\u0026ndash;1612\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePettersen EF et al (2021) UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci 30:70\u0026ndash;82\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"6269f5e5-69a5-4cc2-8bf2-3f9defc58c62","identifier":"10.13039/501100006769","name":"Russian Science Foundation","awardNumber":"23-74-00062","order_by":0}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"prions, amyloids, Sup35, cryogenic electron microscopy","lastPublishedDoi":"10.21203/rs.3.rs-8764517/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8764517/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHuman prions and amyloids exhibit structural diversity that correlates with different pathology manifestations. Yeast prions represent a convenient model for studying basic properties of prions and their strain diversity. The yeast Sup35 prion has multiple structural variants divided into two groups, \u0026ldquo;strong\u0026rdquo; and \u0026ldquo;weak\u0026rdquo;, differing in the frequency of fibril fragmentation and resulting nonsense-suppressor phenotype. Yet, the molecular origin of these distinctions remains unclear. Here, we show that the difference in fragmentation between strong- and weak-type amyloid fibrils correlates with their stability, rather than with the efficiency of chaperone binding. The fully protease-resistant part (residues 2\u0026ndash;32) of [\u003cem\u003ePSI+\u003c/em\u003e] fibril core is a key for the prion phenotype and maintenance, whereas partially resistant part (33\u0026ndash;72) stabilizes the \u0026ldquo;strong\u0026rdquo; prion structure, and is less important for the weak variant. Using prion purified from yeast, we established the cryo-EM structure of the strong [\u003cem\u003ePSI+\u003c/em\u003e] variant, comprising residues 2\u0026ndash;64, with 2.7 \u0026Aring; resolution. Using \u003cem\u003ein silico\u003c/em\u003e modeling based on this atomic structure, we demonstrate that known anti-prion mutations in Sup35 N-domain decrease the thermodynamic stability of the structure. Comparison of \u003cem\u003eex vivo\u003c/em\u003e Sup35 prion structure with that of Sup35 amyloids formed \u003cem\u003ein vitro\u003c/em\u003e shows that the latter differ significantly from the prion structure propagating in yeast cells.\u003c/p\u003e","manuscriptTitle":"Cryo-EM structure of ex vivo Sup35 yeast prion and the factors defining prion phenotype","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-03 09:00:01","doi":"10.21203/rs.3.rs-8764517/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e5700a77-a235-4236-8908-a07f114de658","owner":[],"postedDate":"February 3rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":62158695,"name":"Structural Biology"}],"tags":[],"updatedAt":"2026-02-03T09:00:02+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-03 09:00:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8764517","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8764517","identity":"rs-8764517","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}