Quantitative NMR analysis of the mechanism and kinetics of chaperone Hsp104 action on amyloid-β42 aggregation and fibril formation

Abstract

The chaperone Hsp104, a member of the Hsp100/Clp family of translocases, prevents fibril formation of a variety of amyloidogenic peptides in a paradoxically substoichiometric manner. To understand the mechanism whereby Hsp104 inhibits fibril formation, we probed the interaction of Hsp104 with the Alzheimer’s amyloid-β42 (Aβ42) peptide using a variety of biophysical techniques. Hsp104 is highly effective at suppressing the formation of Thioflavin T (ThT) reactive mature fibrils that are readily observed by atomic force (AFM) and electron (EM) microscopies. Quantitative kinetic analysis and global fitting was performed on serially recorded 1 H- 15 N correlation spectra to monitor the disappearance of Aβ42 monomers during the course of aggregation over a wide range of Hsp104 concentrations. Under the conditions employed (50 μM Aβ42 at 20 °C), Aβ42 aggregation occurs by a branching mechanism: an irreversible on-pathway leading to mature fibrils that entails primary and secondary nucleation and saturating elongation; and a reversible off-pathway to form nonfibrillar oligomers, unreactive to ThT and too large to be observed directly by NMR, but too small to be visualized by AFM or EM. Hsp104 binds reversibly with nanomolar affinity to sparsely populated Aβ42 nuclei present in nanomolar concentrations, generated by primary and secondary nucleation, thereby completely inhibiting on-pathway fibril formation at substoichiometric ratios of Hsp104 to Aβ42 monomers. Tight binding to sparsely populated nuclei likely constitutes a general mechanism for substoichiometric inhibition of fibrillization by a variety of chaperones. Hsp104 also impacts off-pathway oligomerization but to a much smaller degree initially reducing and then increasing the rate of off-pathway oligomerization.

A Hairpin Motif in the Amyloid-β Peptide Is Important for Formation of Disease-Related Oligomers

“…It has interestingly been suggested that some chaperone proteins with antiamyloid activity, such as BRICHOS and the Hsp40-type DNAJB6 chaperone, might bind their clients by forming complementary β-strand/β-strand interactions between client and chaperone . DNAJB6 is known to specifically bind Aβ oligomers rather than monomers or fibrils and therefore must be able to discriminate between these structures. , An NMR study has also shown that it is the hairpin-forming CHC and CTR segments in Aβ that become immobilized upon binding to Hsp104 . All this could indicate that the transient β-hairpin motif is a structure that is recognized by important antiamyloid systems in vivo .…”

“… 92 , 93 An NMR study has also shown that it is the hairpin-forming CHC and CTR segments in Aβ that become immobilized upon binding to Hsp104. 94 All this could indicate that the transient β-hairpin motif is a structure that is recognized by important antiamyloid systems in vivo . Such specific binding would inhibit the formation of the frustrated extended oligomers which nucleate into amyloid.…”

Nucleation of Huntingtin Aggregation Proceeds via Conformational Conversion of Pre‐Formed, Sparsely‐Populated Tetramers

“…The temporal changes in the concentration of tetramers T in monomer units, 4[ T ] = 4 f [ m ( t )], occurring during the course of aggregation for htt ex1 Q 35 samples with total htt ex1 Q 35 concentrations m tot of 160, 250, and 420 µ m , are shown in Figure 3A , while the corresponding htt ex1 Q 35 aggregation profiles best‐fit to the model depicted in Figure 1 and described by Equations ( 1 ) and ( 2 ), are shown in Figure 3B . Throughout this work, the decrease in the NMR signal arising from NMR visible htt ex1 Q 35 monomers as a function of time (Figure 3B ) is attributed to the formation of any species (e.g., fibrils but also, if present, off‐pathway large oligomers) [ 26 ] that are NMR invisible (unobservable) due to their very high molecular weight (see “Experimental Section”). The amount of tetrameric species available at the initial stages of fibril formation (early times in Figure 3A ), defines the rate of primary nucleation (and, hence, the overall initial speed of aggregation) and is inversely proportional to the length of the lag period for the onset of aggregation (Figure 3B ).…”

“…[ 17 ] The high‐quality best‐fits of htt ex1 Q 35 aggregation profiles obtained with the minimalistic kinetic model (Figure 3B and Equations ( 1 ) and ( 2 )) fully satisfy the experimental data, and are devoid of any systematic deviations that would serve as an indication of the presence of additional processes not included in the model. In contrast, in our recent NMR‐based study of aggregation kinetics of Aβ42, [ 26 ] the introduction of an additional step for “off‐pathway” oligomer formation to an otherwise similar kinetic model, was critically important to best fit the aggregation profiles quantitatively. We therefore conclude that further complication of the mechanism of htt ex1 Q 35 aggregation (i.e., the incorporation of additional processes) is not warranted by our experimental data.…”

The C-terminal domain of the antiamyloid chaperone DNAJB6 binds to amyloid-β peptide fibrils and inhibits secondary nucleation

“…In case of Alzheimer's disease, associated with the amyloid formation of Aβ peptide herein investigated, recent data from complex in vivo studies in new neuronal model systems suggest lysosomal quality control deficits in diseased neurons ( 41 , 42 ). A number of different chaperones may inhibit amyloid formation ( 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 43 , 44 , 45 , 46 , 47 ). The mechanism of amyloid formation has been extensively studied in vitro for several proteins, revealing a conformity in terms of underlying steps; in principle, for all systems studied, there appears to be primary and secondary nucleation, elongation, and in some cases, also fragmentation.…”