Yeast Sup35 Prion Structure: Two Types, Four Parts, Many Variants.

The yeast [PSI+] prion, formed by the Sup35 (eRF3) protein, has multiple structural variants differing in the strength of nonsense suppressor phenotype. Structure of [PSI+] and its variation are characterized poorly. Here, we mapped Sup35 amyloid cores of 26 [PSI+] ex vivo prions of different origin using proteinase K digestion and mass spectrometric identification of resistant peptides. In all [PSI+] variants the Sup35 amino acid residues 2-32 were fully resistant and the region up to residue 72 was partially resistant. Proteinase K-resistant structures were also found within regions 73-124, 125-153, and 154-221, but their presence differed between [PSI+] isolates. Two distinct digestion patterns were observed for region 2-72, which always correlated with the "strong" and "weak" [PSI+] nonsense suppressor phenotypes. Also, all [PSI+] with a weak pattern were eliminated by multicopy HSP104 gene and were not toxic when combined with multicopy SUP35. [PSI+] with a strong pattern showed opposite properties, being resistant to multicopy HSP104 and lethal with multicopy SUP35. Thus, Sup35 prion cores can be composed of up to four elements. [PSI+] variants can be divided into two classes reliably distinguishable basing on structure of the first element and the described assays.

The Pub1 and Upf1 Proteins Act in Concert to Protect Yeast from Toxicity of the [PSI⁺] Prion.

The [PSI⁺] nonsense-suppressor determinant of Saccharomyces cerevisiae is based on the formation of heritable amyloids of the Sup35 (eRF3) translation termination factor. [PSI⁺] amyloids have variants differing in amyloid structure and in the strength of the suppressor phenotype. The appearance of [PSI⁺], its propagation and manifestation depend primarily on chaperones. Besides chaperones, the Upf1/2/3, Siw14 and Arg82 proteins restrict [PSI⁺] formation, while Sla2 can prevent [PSI⁺] toxicity. Here, we identify two more non-chaperone proteins involved in [PSI⁺] detoxification. We show that simultaneous lack of the Pub1 and Upf1 proteins is lethal to cells harboring [PSI⁺] variants with a strong, but not with a weak, suppressor phenotype. This lethality is caused by excessive depletion of the Sup45 (eRF1) termination factor due to its sequestration into Sup35 polymers. We also show that Pub1 acts to restrict excessive Sup35 prion polymerization, while Upf1 interferes with Sup45 binding to Sup35 polymers. These data allow consideration of the Pub1 and Upf1 proteins as a novel [PSI⁺] detoxification system.

Distinct mechanisms of mutant huntingtin toxicity in different yeast strains.

Expansion of polyglutamine stretches in several proteins causes neurodegenerative amyloidoses, including Huntington disease. In yeast, mutant huntingtin (mHtt) with a stretch of 103 glutamine residues (HttQ103) forms toxic aggregates. A range of yeast strains have been used to elucidate the mechanisms of mHtt toxicity, and have revealed perturbations of various unrelated processes. HttQ103 aggregates can induce aggregation of cellular proteins, many of which contain glutamine/asparagine-rich regions, including Sup35 and Def1. In the strain 74-D694 HttQ103, toxicity is related to aggregation-mediated depletion of soluble Sup35 and its interacting partner Sup45. Def1 was also implicated in mHtt toxicity, since its lack detoxified HttQ103 in another yeast strain, BY4741. Here we show that in BY4742, deletion of DEF1 lowers HttQ103 toxicity and decreases the amount of its polymers, but does not affect copolymerization of Sup35. Furthermore, in contrast to 74-D694, increasing the levels of soluble Sup35 and Sup45 does not alleviate toxicity of HttQ103 in BY4742. These data demonstrate a difference in the mechanisms underlying mHtt toxicity in different yeast strains and suggest that in humans with Huntington disease, neurons of different brain compartments and cells in other tissues can also be damaged by different mechanisms.

Molecular basis for transmission barrier and interference between closely related prion proteins in yeast.

Replicating amyloids, called prions, are responsible for transmissible neurodegenerative diseases in mammals and some heritable phenotypes in fungi. The transmission of prions between species is usually inhibited, being highly sensitive to small differences in amino acid sequence of the prion-forming proteins. To understand the molecular basis of this prion interspecies barrier, we studied the transmission of the [PSI(+)] prion state from Sup35 of Saccharomyces cerevisiae to hybrid Sup35 proteins with prion-forming domains from four other closely related Saccharomyces species. Whereas all the hybrid Sup35 proteins could adopt a prion form in S. cerevisiae, they could not readily acquire the prion form from the [PSI(+)] prion of S. cerevisiae. Expression of the hybrid Sup35 proteins in S. cerevisiae [PSI(+)] cells often resulted in frequent loss of the native [PSI(+)] prion. Furthermore, all hybrid Sup35 proteins showed different patterns of interaction with the native [PSI(+)] prion in terms of co-polymerization, acquisition of the prion state, and induced prion loss, all of which were also dependent on the [PSI(+)] variant. The observed loss of S. cerevisiae [PSI(+)] can be related to inhibition of prion polymerization of S. cerevisiae Sup35 and formation of a non-heritable form of amyloid. We have therefore identified two distinct molecular origins of prion transmission barriers between closely sequence-related prion proteins: first, the inability of heterologous proteins to co-aggregate with host prion polymers, and second, acquisition by these proteins of a non-heritable amyloid fold.

Elongation factor eEF1B modulates functions of the release factors eRF1 and eRF3 and the efficiency of translation termination in yeast.

BACKGROUND: Termination of translation in eukaryotes is controlled by two interacting polypeptide chain release factors, eRF1 and eRF3. While eRF1 recognizes nonsense codons, eRF3 facilitates polypeptide chain release from the ribosome in a GTP-dependent manner. Besides termination, both release factors have essential, but poorly characterized functions outside of translation. RESULTS: To characterize further the functions of yeast eRF1 and eRF3, a genetic screen for their novel partner proteins was performed. As a result, the genes for gamma (TEF4 and TEF3/CAM1) and alpha (TEF5/EFB1) subunits of the translation elongation factor eEF1B, known to catalyze the exchange of bound GDP for GTP on eEF1A, were revealed. These genes act as dosage suppressors of a synthetic growth defect caused by some mutations in the SUP45 and SUP35 genes encoding eRF1 and eRF3, respectively. Extra copies of TEF5 and TEF3 can also suppress the temperature sensitivity of some sup45 and sup35 mutants and reduce nonsense codon readthrough caused by these omnipotent suppressors. Besides, overproduction of eEF1Balpha reduces nonsense codon readthrough in the strain carrying suppressor tRNA. Such effects were not shown for extra copies of TEF2, which encodes eEF1A, thus indicating that they were not due to eEF1A activation. CONCLUSION: The data obtained demonstrate involvement of the translation elongation factor eEF1B in modulating the functions of translation termination factors and suggest its possible role in GDP for GTP exchange on eRF3.

Analysis of novel hyperosmotic shock response suggests 'beads in liquid' cytosol structure.

Proteins can aggregate in response to stresses, including hyperosmotic shock. Formation and disassembly of aggregates is a relatively slow process. We describe a novel instant response of the cell to hyperosmosis, during which chaperones and other proteins form numerous foci with properties uncharacteristic of classical aggregates. These foci appeared/disappeared seconds after shock onset/removal, in close correlation with cell volume changes. Genome-wide and targeted testing revealed chaperones, metabolic enzymes, P-body components and amyloidogenic proteins in the foci. Most of these proteins can form large assemblies and for some, the assembled state was pre-requisite for participation in foci. A genome-wide screen failed to identify genes whose absence prevented foci participation by Hsp70. Shapes of and interconnections between foci, revealed by super-resolution microscopy, indicated that the foci were compressed between other entities. Based on our findings, we suggest a new model of cytosol architecture as a collection of numerous gel-like regions suspended in a liquid network. This network is reduced in volume in response to hyperosmosis and forms small pockets between the gel-like regions.

Dissection of differential vanadate sensitivity in two Ogataea species links protein glycosylation and phosphate transport regulation.

The closely related yeasts Ogataea polymorpha and O. parapolymorpha differ drastically from each other by sensitivity to the toxic phosphate analog vanadate. Search for genes underlying this difference revealed two genes, one designated as ABV1 (Alcian Blue staining, Vanadate resistance), which encodes a homologue of Saccharomyces cerevisiae Mnn4 responsible for attachment of mannosylphosphate to glycoside chains of secretory proteins, and the other designated as its S. cerevisiae homologue PHO87, encoding the plasma membrane low affinity phosphate sensor/transporter. The effect of Pho87 on vanadate resistance was bidirectional, since it decreased the resistance on phosphate-depleted medium, but was required for pronounced protection against vanadate by external phosphate. This highlights the dual function of this protein as a low affinity phosphate transporter and an external phosphate sensor. Involvement of Pho87 in phosphate sensing was confirmed by its effects on regulation of the promoter of the PHO84 gene, encoding a high affinity phosphate transporter. The effect of Abv1 was also complex, since it influenced Pho87 level and enhanced repression of the PHO84 promoter via a Pho87-independent pathway. Role of the identified genes in the difference in vanadate resistance between O. polymorpha and O. parapolymorpha is discussed.

The role of the N-terminal oligopeptide repeats of the yeast Sup35 prion protein in propagation and transmission of prion variants.

The cytoplasmic [PSI+] determinant of Saccharomyces cerevisiae is the prion form of the Sup35 protein. Oligopeptide repeats within the Sup35 N-terminal domain (PrD) presumably are required for the stable [PSI+] inheritance that in turn involves fragmentation of Sup35 polymers by the chaperone Hsp104. The nonsense suppressor [PSI+] phenotype can vary in efficiency probably due to different inheritable Sup35 polymer structures. Here we study the ability of Sup35 mutants with various deletions of the oligopeptide repeats to support [PSI+] propagation. We define the minimal region of the Sup35-PrD necessary to support [PSI+] as amino acids 1-64, which include the first two repeats, although a longer fragment, 1-83, is required to maintain weak [PSI+] variants. Replacement of wild-type Sup35 with deletion mutants decreases the strength of the [PSI+] phenotype. However, with one exception, reintroducing the wild-type Sup35 restores the original phenotype. Thus, the specific prion fold defining the [PSI+] variant can be preserved by the mutant Sup35 protein despite the change of phenotype. Coexpression of wild-type and mutant Sup35 containing three, two, one, or no oligopeptide repeats causes variant-specific [PSI+] elimination. These data suggest that [PSI+] variability is primarily defined by differential folding of the Sup35-PrD oligopeptide-repeat region.

Ribosome-bound Pub1 modulates stop codon decoding during translation termination in yeast.

In eukaryotes, termination of translation is controlled by polypeptide chain release factors eRF1 and eRF3, of which the former recognizes nonsense codons, while the latter interacts with eRF1 and stimulates polypeptide release from the ribosome in a GTP- dependent manner, and ABCE1, which facilitates ribosome recycling. In this work, we demonstrate that Pub1, a yeast protein known to be involved in stress granule formation, regulation of gene expression, and organization of the tubulin cytoskeleton, also plays a role in translation termination. Pub1 was shown to bind to ribosomes independent of eRF1 and eRF3 and to interact with the N-terminal glutamine-/asparagine-rich prion domain of eRF3 via its short C-terminal glutamine-rich tract. High velocity sedimentation in sucrose gradient demonstrated that Pub1 was preferentially associated with heavy polysomes enriched with terminating ribosomes. Lack of Pub1 decreased efficiency of nonsense readthrough at a majority but not all tetranucleotide stop signals. This distinguishes Pub1 from most other known binding partners of the release factors which were shown to modulate readthrough of all nonsense codons uniformly. The obtained data show that Pub1 can act as an accessory translation factor involved in fine-tuning of translation termination.

N-terminal region of Saccharomyces cerevisiae eRF3 is essential for the functioning of the eRF1/eRF3 complex beyond translation termination.

BACKGROUND: Termination of translation in eukaryotes requires two release factors, eRF1, which recognizes all three nonsense codons and facilitates release of the nascent polypeptide chain, and eRF3 stimulating translation termination in a GTP-depended manner. eRF3 from different organisms possess a highly conservative C region (eRF3C), which is responsible for the function in translation termination, and almost always contain the N-terminal extension, which is inessential and vary both in structure and length. In the yeast Saccharomyces cerevisiae the N-terminal region of eRF3 is responsible for conversion of this protein into the aggregated and functionally inactive prion form. RESULTS: Here, we examined functional importance of the N-terminal region of a non-prion form of yeast eRF3. The screen for mutations which are lethal in combination with the SUP35-C allele encoding eRF3C revealed the sup45 mutations which alter the N-terminal domain of eRF1 and increase nonsense codon readthrough. However, further analysis showed that synthetic lethality was not caused by the increased levels of nonsense codon readthrough. Dominant mutations in SUP35-C were obtained and characterized, which remove its synthetic lethality with the identified sup45 mutations, thus indicating that synthetic lethality was not due to a disruption of interaction with proteins that bind to this eRF3 region. CONCLUSION: These and other data demonstrate that the N-terminal region of eRF3 is involved both in modulation of the efficiency of translation termination and functioning of the eRF1/eRF3 complex outside of translation termination.

Novel non-Mendelian determinant involved in the control of translation accuracy in Saccharomyces cerevisiae.

Two cytoplasmically inherited determinants related by their manifestation to the control of translation accuracy were previously described in yeast. Cells carrying one of them, [PSI(+)], display a nonsense suppressor phenotype and contain a prion form of the Sup35 protein. Another element, [PIN(+)], determines the probability of de novo generation of [PSI(+)] and results from a prion form of several proteins, which can be functionally unrelated to Sup35p. Here we describe a novel nonchromosomal determinant related to the SUP35 gene. This determinant, designated [ISP(+)], was identified as an antisuppressor of certain sup35 mutations. We observed its loss upon growth on guanidine hydrochloride and subsequent spontaneous reappearance with high frequency. The reversible curability of [ISP(+)] resembles the behavior of yeast prions. However, in contrast to known prions, [ISP(+)] does not depend on the chaperone protein Hsp104. Though manifestation of both [ISP(+)] and [PSI(+)] is related to the SUP35 gene, the maintenance of [ISP(+)] does not depend on the prionogenic N-terminal domain of Sup35p and Sup35p is not aggregated in [ISP(+)] cells, thus ruling out the possibility that [ISP(+)] is a specific form of [PSI(+)]. We hypothesize that [ISP(+)] is a novel prion involved in the control of translation accuracy in yeast.

Genetic Evidence for the Role of the Vacuole in Supplying Secretory Organelles with Ca2+ in Hansenula polymorpha.

Processes taking place in the secretory organelles require Ca2+ and Mn2+, which in yeast are supplied by the Pmr1 ion pump. Here we observed that in the yeast Hansenula polymorpha Ca2+ deficiency in the secretory pathway caused by Pmr1 inactivation is exacerbated by (i) the ret1-27 mutation affecting COPI-mediated vesicular transport, (ii) inactivation of the vacuolar Ca2+ ATPase Pmc1 and (iii) inactivation of Vps35, which is a component of the retromer complex responsible for protein transport between the vacuole and secretory organelles. The ret1-27 mutation also exerted phenotypes indicating alterations in transport between the vacuole and secretory organelles. These data indicate that ret1-27, pmc1 and vps35 affect a previously unknown Pmr1-independent route of the Ca2+ delivery to the secretory pathway. We also observed that the vacuolar protein carboxypeptidase Y receives additional modifications of its glycoside chains if it escapes the Vps10-dependent sorting to the vacuole.

A protein polymerization cascade mediates toxicity of non-pathological human huntingtin in yeast.

Several neurodegenerative amyloidoses, including Huntington disease, are caused by expansion of polyglutamine (polyQ) stretches in otherwise unrelated proteins. In a yeast model, an N-terminal fragment of mutant huntingtin with a stretch of 103 glutamine residues aggregates and causes toxicity, while its non-toxic wild type variant with a sequence of 25 glutamines (Htt25Q) does not aggregate. Here, we observed that non-toxic polymers of various proteins with glutamine-rich domains could seed polymerization of Htt25Q, which caused toxicity by seeding polymerization of the glutamine/asparagine-rich Sup35 protein thus depleting the soluble pools of this protein and its interacting partner, Sup45. Importantly, only polymers of Htt25Q, but not of the initial benign polymers, induced Sup35 polymerization, indicating an intermediary role of Htt25Q in cross-seeding Sup35 polymerization. These data provide a novel insight into interactions between amyloidogenic proteins and suggest a possible role for these interactions in the pathogenesis of Huntington and other polyQ diseases.

Aggregation and retention of human urokinase type plasminogen activator in the yeast endoplasmic reticulum.

BACKGROUND: Secretion of recombinant proteins in yeast can be affected by their improper folding in the endoplasmic reticulum and subsequent elimination of the misfolded molecules via the endoplasmic reticulum associated protein degradation pathway. Recombinant proteins can also be degraded by the vacuolar protease complex. Human urokinase type plasminogen activator (uPA) is poorly secreted by yeast but the mechanisms interfering with its secretion are largely unknown. RESULTS: We show that in Hansenula polymorpha overexpression worsens uPA secretion and stimulates its intracellular aggregation. The absence of the Golgi modifications in accumulated uPA suggests that aggregation occurs within the endoplasmic reticulum. Deletion analysis has shown that the N-terminal domains were responsible for poor uPA secretion and propensity to aggregate. Mutation abolishing N-glycosylation decreased the efficiency of uPA secretion and increased its aggregation degree. Retention of uPA in the endoplasmic reticulum stimulates its aggregation. CONCLUSIONS: The data obtained demonstrate that defect of uPA secretion in yeast is related to its retention in the endoplasmic reticulum. Accumulation of uPA within the endoplasmic reticulum disturbs its proper folding and leads to formation of high molecular weight aggregates.

Correct GPI-anchor synthesis is required for the incorporation of endoglucanase/glucanosyltransferase Bgl2p into the Saccharomyces cerevisiae cell wall.

The SSU21/MCD4 gene encodes an essential component of the glycosylphosphatidylinositol (GPI)-anchor synthesis pathway in Saccharomyces cerevisiae. Here we demonstrate that the ssu21 mutation affected the transport and the incorporation into the cell wall of the major non-GPI yeast cross-linker - endoglucanase/glucanosyltransferase Bgl2p. This mutation also led to a decrease in the levels of both known types of cell wall mannoproteins, those covalently linked with glucan and SDS-extractable proteins. Our results indicate that the precision of the GPI-anchor synthesis is essential for cell wall assembly and suggest the strong interdependence of different groups of cell wall proteins during their incorporation into the cell wall.

Proteomic screening for amyloid proteins.

Despite extensive study, progress in elucidation of biological functions of amyloids and their role in pathology is largely restrained due to the lack of universal and reliable biochemical methods for their discovery. All biochemical methods developed so far allowed only identification of glutamine/asparagine-rich amyloid-forming proteins or proteins comprising amyloids that form large deposits. In this article we present a proteomic approach which may enable identification of a broad range of amyloid-forming proteins independently of specific features of their sequences or levels of expression. This approach is based on the isolation of protein fractions enriched with amyloid aggregates via sedimentation by ultracentrifugation in the presence of strong ionic detergents, such as sarkosyl or SDS. Sedimented proteins are then separated either by 2D difference gel electrophoresis or by SDS-PAGE, if they are insoluble in the buffer used for 2D difference gel electrophoresis, after which they are identified by mass-spectrometry. We validated this approach by detection of known yeast prions and mammalian proteins with established capacity for amyloid formation and also revealed yeast proteins forming detergent-insoluble aggregates in the presence of human huntingtin with expanded polyglutamine domain. Notably, with one exception, all these proteins contained glutamine/asparagine-rich stretches suggesting that their aggregates arose due to polymerization cross-seeding by human huntingtin. Importantly, though the approach was developed in a yeast model, it can easily be applied to any organism thus representing an efficient and universal tool for screening for amyloid proteins.

Could yeast prion domains originate from polyQ/N tracts?

A significant body of evidence shows that polyglutamine (polyQ) tracts are important for various biological functions. The characteristic polymorphism of polyQ length is thought to play an important role in the adaptation of organisms to their environment. However, proteins with expanded polyQ are prone to form amyloids, which cause diseases in humans and animals and toxicity in yeast. Saccharomyces cerevisiae contain at least 8 proteins which can form heritable amyloids, called prions, and most of them are proteins with glutamine- and asparagine-enriched domains. Yeast prion amyloids are susceptible to fragmentation by the protein disaggregase Hsp104, which allows them to propagate and be transmitted to daughter cells during cell divisions. We have previously shown that interspersion of polyQ domains with some non-glutamine residues stimulates fragmentation of polyQ amyloids in yeast and that yeast prion domains are often enriched in one of these residues. These findings indicate that yeast prion domains may have derived from polyQ tracts via accumulation and amplification of mutations. The same hypothesis may be applied to polyasparagine (polyN) tracts, since they display similar properties to polyQ, such as length polymorphism, amyloid formation and toxicity. We propose that mutations in polyQ/N may be favored by natural selection thus making prion domains likely by-products of the evolution of polyQ/N.

Amyloid-mediated sequestration of essential proteins contributes to mutant huntingtin toxicity in yeast.

BACKGROUND: Polyglutamine expansion is responsible for several neurodegenerative disorders, among which Huntington disease is the most well-known. Studies in the yeast model demonstrated that both aggregation and toxicity of a huntingtin (htt) protein with an expanded polyglutamine region strictly depend on the presence of the prion form of Rnq1 protein ([PIN+]), which has a glutamine/asparagine-rich domain. PRINCIPAL FINDINGS: Here, we showed that aggregation and toxicity of mutant htt depended on [PIN+] only quantitatively: the presence of [PIN+] elevated the toxicity and the levels of htt detergent-insoluble polymers. In cells lacking [PIN+], toxicity of mutant htt was due to the polymerization and inactivation of the essential glutamine/asparagine-rich Sup35 protein and related inactivation of another essential protein, Sup45, most probably via its sequestration into Sup35 aggregates. However, inhibition of growth of [PIN+] cells depended on Sup35/Sup45 depletion only partially, suggesting that there are other sources of mutant htt toxicity in yeast. CONCLUSIONS: The obtained data suggest that induced polymerization of essential glutamine/asparagine-rich proteins and related sequestration of other proteins which interact with these polymers represent an essential source of htt toxicity.

Inactivation of Pmc1 vacuolar Ca(2+) ATPase causes G(2) cell cycle delay in Hansenula polymorpha.

The vacuolar Ca(2+) ATPase Pmc1 is involved in maintenance of a low Ca(2+) concentration in cytosol in yeast cells. Here we observed that increase of Ca(2+) cytosolic concentration in yeast Hansenula polymorpha due to inactivation of Pmc1 resulted in sensitivity to sodium dodecyl sulfate (SDS). To elucidate the mechanisms of the observed effect, a screening for mutations suppressing SDS sensitivity of the H. polymorpha pmc1 mutant was performed. As a result, three genes were identified. Two of them, designated as their Saccharomyces cerevisiae orthologs CCH1 and HOG1 encoded the plasma membrane voltage-gated high-affinity calcium channel and the MAP kinase involved in osmoregulation, respectively. The third gene, designated as WEE1, coded for the ortholog of Wee1/Swe1 kinase involved in cell cycle regulation by inhibiting of the G(2)/M transition. Detailed analysis of this mutant demonstrated that suppression of pmc1 SDS sensitivity by the wee1 mutation depended on an accompanying chromosomal rearrangement, whereas inactivation of WEE1 in the absence of this rearrangement caused SDS sensitivity. Expression of a chimeric protein containing an N-terminal portion of Wee1 in the pmc1 mutant led to abnormal morphology characteristic of G(2) delay. Our data indicate that cytosolic Ca(2+) rise causes SDS sensitivity in H. polymorpha through the activation of the Wee1 kinase, which is mediated by the Hog1 kinase. Wee1 has a dual role in the manifestation of SDS sensitivity in the H. polymorpha pmc1 mutant. Mechanisms of influence of the obtained mutations on the G(2)/M transition are discussed.

The effects of amino acid composition of glutamine-rich domains on amyloid formation and fragmentation.

Fragmentation of amyloid polymers by the chaperone Hsp104 allows them to propagate as prions in yeast. The factors which determine the frequency of fragmentation are unclear, though it is often presumed to depend on the physical strength of prion polymers. Proteins with long polyglutamine stretches represent a tractable model for revealing sequence elements required for polymer fragmentation in yeast, since they form poorly fragmented amyloids. Here we show that interspersion of polyglutamine stretches with various amino acid residues differentially affects the in vivo formation and fragmentation of the respective amyloids. Aromatic residues tyrosine, tryptophan and phenylalanine strongly stimulated polymer fragmentation, leading to the appearance of oligomers as small as dimers. Alanine, methionine, cysteine, serine, threonine and histidine also enhanced fragmentation, while charged residues, proline, glycine and leucine inhibited polymerization. Our data indicate that fragmentation frequency primarily depends on the recognition of fragmentation-promoting residues by Hsp104 and/or its co-chaperones, rather than on the physical stability of polymers. This suggests that differential exposure of such residues to chaperones defines prion variant-specific differences in polymer fragmentation efficiency.