Amyloid-forming Peptides from β2-Microglobulin—Insights into the Mechanism of Fibril Formation in Vitro

Jones, Susan R.; Manning, James E.; Kad, Neil M.; Radford, Sheena E.

doi:10.1016/s0022-2836(02)01227-5

Cited by 144 publications

(160 citation statements)

References 49 publications

Supporting

Mentioning

146

Contrasting

Unclassified

Order By: Relevance

“…Our scanning predicts that strand E is indeed the only amyloidogenic fragment of ␤ 2 m. Consistent with this, out of all synthesized molecules comprising ␤ 2 m strands, only peptides containing the sequence of strand E form amyloids (36).…”

Section: Avoidance Of Amyloid Fibril Formation In Proteins: Amyloid Bsupporting

confidence: 82%

Sequence determinants of amyloid fibril formation

Paz

Serrano²

2003

Proc. Natl. Acad. Sci. U.S.A.

370

360

View full text Add to dashboard Cite

The establishment of rules that link sequence and amyloid feature is critical for our understanding of misfolding diseases. To this end, we have performed a saturation mutagenesis analysis on the de novodesigned amyloid peptide STVIIE (1). The positional scanning mutagenesis has revealed that there is a position dependence on mutation of amyloid fibril formation and that both very tolerant and restrictive positions to mutation can be found within an amyloid sequence. In this system, mutations that accelerate ␤-sheet polymerization do not always lead to an increase of amyloid products. On the contrary, abundant fibrils are typically found for mutants that polymerize slowly. From these experiments, we have extracted a sequence pattern to identify amyloidogenic stretches in proteins. The pattern has been validated experimentally. In silico sequence scanning of amyloid proteins also supports the pattern. Analysis of protein databases has shown that highly amyloidogenic sequences matching the pattern are less frequent in proteins than innocuous amino acid combinations and that, if present, they are surrounded by amino acids that disrupt their aggregating capability (amyloid breakers). This study provides the potential for a proteome-wide scanning to detect fibril-forming regions in proteins, from which molecules can be designed to prevent and͞or disrupt this process.I t is accepted that misfolding of peptides and proteins cause the fibrillar aggregates that characterize the group of diseases known as amyloidoses (1, 2). Despite active research, a detailed understanding of the molecular principles underlying the transformation of soluble proteins into amyloid aggregates is still lacking (1-3).The group of peptides and proteins capable of forming amyloid fibrils is very diverse (4-8). This group does not only consist of proteins involved in amyloid deposits in vivo (4), but also of nonpathogenic (15, 16) and designed peptides and proteins (7,8). X-ray fiber diffraction data indicate that all amyloid fibrils share a cross-␤-structure, regardless of the sequence or native fold of the soluble precursor (1, 2). Thus, an increasingly adopted view is that the ability to form amyloid fibrils is a general property of the polypeptide backbone (6). Nevertheless, the propensity of a given polypeptide to form amyloid fibrils depends enormously on amino acid composition (7-10).A protein has to be partially or fully unfolded to aggregate (1, 2). Yet, mutations leading to aggregation can alter (9, 10), or not alter (11, 12), protein stability. Aggregation from some peptides and proteins involved in relevant amyloidoses, such as type II diabetes and Alzheimer's and Parkinson's disease, does not require, however, previous unfolding, because these molecules are largely unstructured under physiological conditions (2). Nevertheless, most of the natively unfolded proteins in vivo do not undergo aggregation (13), indicating that unfolding is necessary, but not sufficient, to promote aggregation. Hence, there must be some sequence motifs tha...

show abstract

Section: Avoidance Of Amyloid Fibril Formation In Proteins: Amyloid Bsupporting

confidence: 82%

Sequence determinants of amyloid fibril formation

Paz

Serrano²

2003

Proc. Natl. Acad. Sci. U.S.A.

370

360

View full text Add to dashboard Cite

show abstract

“…Two peptides (residues 21-40, the so-called K3 peptide, and residues 59-71), which overlap with the regions that are strongly exchange broadened in the pH 3.6 intermediate, were reported to self-associate in Vitro. 22,23 The two -strands (residues 21-28 and 33-40) observed in amyloid fibrils of the K3 peptide 24 match the two minima in the H/D exchange profile (Figure 2), suggesting that amyloid fibrils of full-length b2m also contain two -strands in this region.We demonstrated that signals of partially unfolded intermediate ensembles broadened due to conformational exchange can be structurally characterized using direct carbon-detected NMR experiments. Comparison between the structural and dynamic properties of the amyloid intermediate of b2m with H/D exchange measurements on amyloid fibrils revealed a close relationship between the conformational properties of the metastable partially unfolded precursor and the -sheet-rich insoluble aggregates of a diseaserelevant protein that assumes a rigid 3D fold in its native state.…”

mentioning

confidence: 67%

Molten Globule Precursor States Are Conformationally Correlated to Amyloid Fibrils of Human β-2-Microglobulin

2010

View full text Add to dashboard Cite

Misfolding intermediates play a key role in defining aberrant protein aggregation and amyloid formation in more than 15 different human diseases. However, their experimental characterization is challenging due to the transient nature and conformational heterogeneity of the involved states. Here, we demonstrate that direct carbon-detected NMR experiments allow observation, assignment, and structural analysis of molten globule amyloid intermediates that are severely broadened by conformational exchange. The method is used to characterize the structure and dynamics of partially unfolded intermediates of the 99-residue protein -2-microglobulin, which is the major component of insoluble aggregates occurring in dialysis-related amyloidosis. Comparison of the conformational properties of the molten globule-like intermediates with levels of deuterium incorporation into amyloid fibrils of -2-microglobulin revealed a close relationship between the conformational properties of the metastable intermediates and the -sheet-rich insoluble aggregates of -2-microglobulin.Misfolding intermediates play a key role in defining aberrant protein aggregation and amyloid formation in more than 15 different human diseases. 1,2 However, the experimental characterization of the conformation of amyloid intermediates is challenging due to their transient nature and conformational heterogeneity. 3,4 This severely limits our knowledge about the relation of the conformation of the precursor states to the structure of amyloid fibrils.Dialysis-related amyloidosis is a protein misfolding disease resulting from deposition of amyloid aggregates in skeletal tissue that contain fibrils of the 99-residue-long protein -2-microglobulin (b2m). 5 Amyloid formation of b2m is strongly enhanced in conditions that destabilize its globular structure. 6 This can be achieved in Vitro by acid denaturation, with two distinct intermediate states being formed under acidic conditions. The highest population of the partially unfolded intermediate occurs at pH 3.6, where also the rate of fibril formation reaches a maximum. 7 However, long and straight amyloid fibrils resembling those extracted from patients are formed from a precursor state formed at pH 2.5. 8,9 Here, we compare the conformational properties of these two partially unfolded intermediates with single-residue resolution and demonstrate that a close relationship exists between the structure and dynamics of the amyloid precursor species and the morphology of mature fibrils of b2m.At pH 2.5 b2m is highly unfolded, the majority of resonances are observed in 1 H-15 N HSQC spectra (Supporting Information, Figure 1) and can be assigned to individual residues. 10 In contrast, only about 40-60 resolved peaks on top of a broad hump of unresolved signal intensity (visible at lower contour levels) could be observed at pH 3.6 ( Figure 1a). 11 To obtain sequence-specific information about the pH 3.6 intermediate, we tested direct 13 C detection. 12-14 Direct carbon detection was previously successfully used fo...

show abstract

“…However, these regulatory mechanisms would be ineffective in a deficient assembly process. Moreover, the presence of a protein subunit free from its parent complex can have serious consequences, because the exposure of the surface normally hidden in the oligomer may trigger an aggregation process (16). For instance, patients on long-term hemodialysis can develop amyloidosis generated by ␤2-microglobulin free from a class I major histocompatibility complex heavy chain (17).…”

Section: Resultsmentioning

confidence: 99%

Glycoprotein Tertiary and Quaternary Structures Are Monitored by the Same Quality Control Mechanism

Keith

Parodi

Caramelo

2005

Journal of Biological Chemistry

View full text Add to dashboard Cite

Folding of glycoproteins entering the secretory pathway is strictly surveyed in the endoplasmic reticulum by a quality control system. Folding intermediates and proteins irreparably misfolded are marked via glucosylation by the UDPglucose:glycoprotein glucosyltransferase, an enzyme that acts as a folding sensor by exclusively labeling glycoproteins not displaying their native structures. Here we show that this sensing mechanism also applies to the oligomerization of protein complexes, as the glucosyltransferase appeared to be able to glucosylate folded complex subunits lacking the full complement of oligomer components.Newly synthesized proteins must complete folding before they move through the secretory pathway. A quality control mechanism retains misfolded and unassembled species in the endoplasmic reticulum (ER) 1 (1, 2). This system is composed of four main elements, the ER resident lectins calnexin and calreticulin and the enzymes glucosidase II and UDPglucose:glycoprotein glucosyltransferase (GT). About 80% of proteins entering the secretory pathway are N-glycosylated cotranslationally in the consensus sequence Asn-X-(Ser/Thr). The two outermost glucoses are removed from the transferred glycan (Glc 3 Man 9 GlcNAc 2 ) by the sequential action of glucosidase I and glucosidase II. The monoglucosylated species thus generated are then recognized by calnexin and/or calreticulin, which are lectins specific for Glc 1 Man 5-9 GlcNAc 2 glycans (3). The glycoprotein-lectin association not only retains in the ER the species that are not properly folded but also enhances folding efficiency by preventing the aggregation of intermediates and allowing the action of other folding-enhancing proteins such as ERp57, a protein disulfide isomerase associated with calnexin and calreticulin (4). The glycoprotein-lectin interaction is eventually terminated upon glucosidase II cleavage of the remaining glucose. At this stage, glycoproteins that are properly folded proceed to their final destination. Folding intermediates or irreparably misfolded glycoproteins, however, are reglucosylated by GT, an enzyme that recreates the monoglucosylated structure, and are thus retained by the lectins in the ER. GT operates as a folding sensor, because it only glucosylates glycoproteins not displaying their native structures (5). Protein determinants triggering GT-mediated glucosylation have been identified as patches of hydrophobic amino acids that are mainly solvent-exposed in molten globule-like conformations (6).It has been observed in vivo that only fully assembled oligomeric proteins leave the ER. The above mentioned glycoprotein folding quality control mechanism has been implicated in the retention of individual subunits or partially assembled oligomers, because those species have been isolated as being associated to calnexin/calreticulin from cells unable to form the complete oligomers (7,8). For instance, the lectin-glycoprotein association of T cell receptor or human class I major histocompatibility complex components has been observed...

show abstract

Amyloid-forming Peptides from β2-Microglobulin—Insights into the Mechanism of Fibril Formation in Vitro

Cited by 144 publications

References 49 publications

Sequence determinants of amyloid fibril formation

Sequence determinants of amyloid fibril formation

Molten Globule Precursor States Are Conformationally Correlated to Amyloid Fibrils of Human β-2-Microglobulin

Glycoprotein Tertiary and Quaternary Structures Are Monitored by the Same Quality Control Mechanism

Contact Info

Product

Resources

About