Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Dissection of the amyloid formation pathway in AL amyloidosis

Dissection of the amyloid formation pathway in AL amyloidosis ARTICLE https://doi.org/10.1038/s41467-021-26845-0 OPEN Dissection of the amyloid formation pathway in AL amyloidosis 1,3 1,3 2 1 Pamina Kazman , Ramona M. Absmeier , Harald Engelhardt & Johannes Buchner In antibody light chain (AL) amyloidosis, overproduced light chain (LC) fragments accumu- late as fibrils in organs and tissues of patients. In vitro, AL fibril formation is a slow process, characterized by a pronounced lag phase. The events occurring during this lag phase are largely unknown. We have dissected the lag phase of a patient-derived LC truncation and identified structural transitions that precede fibril formation. The process starts with partial unfolding of the V domain and the formation of small amounts of dimers. This is a pre- requisite for the formation of an ensemble of oligomers, which are the precursors of fibrils. During oligomerization, the hydrophobic core of the LC domain rearranges which leads to changes in solvent accessibility and rigidity. Structural transitions from an anti-parallel to a parallel β-sheet secondary structure occur in the oligomers prior to amyloid formation. Together, our results reveal a rate-limiting multi-step mechanism of structural transitions prior to fibril formation in AL amyloidosis, which offers, in the long run, opportunities for therapeutic intervention. 1 2 Department Chemie, Technische Universität München, 85748 Garching, Germany. Department Molecular Structural Biology, Max-Planck-Institute of 3 ✉ Biochemistry, 82152 Martinsried, Germany. These authors contributed equally: Pamina Kazman, Ramona M. Absmeier. email: johannes.buchner@tum.de NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 16,23 ight chain (AL) amyloidosis is the most common type of cooperative reaction as monitored by ThT fluorescence .The systemic amyloidosis . The disease is caused by an under- molecular events occurring in this lag phase are still elusive. For a Llying plasma cell dyscrasia that entails the elevated expres- better understanding of AL amyloidosis, we set out to resolve the 2,3 sion and secretion of free antibody light chains (LC) .Patient- molecular mechanism of structural changes occurring during the lag specificmutations in theLC, whichdestabilize thenativefoldand phase for the well-characterized patient LC truncation Pat-1 that consequently trigger fibril formation are an important element of the represents the major component of the deposited fibrils. The struc- 4–7 disease . Furthermore, the circulating free LCs often undergo ture and amyloidogenic properties of Pat-1 had been reported proteolytic cleavage prior to assembly into insoluble amyloid previously . In this study, two disease-causing mutations (L15P 8–11 fibrils .N-terminalfragments,comprisingmostlythe variable L82Q) had been identified in Pat-1 compared to the germline domain (V ) of the LC often represent the amyloidogenic species. sequence, which did not form fibrils. For the lag phase analysis, we The hallmark of AL amyloidosis is the transformation of soluble used the Pat-1 V , the double mutant Pat-1 L15P L82Q and the 12,13 monomeric protein into insoluble amyloid fibrils . Their presence respective germline sequence WT-1. To obtain insights into the correlates strongly with the diseaseand theimpairmentoforgansin molecular events occurring during the lag phase, we first monitored which they are deposited. Structurally, substantial rearrangements of changes in the solubility of the V domain under conditions favoring the V domain in AL fibrils compared to the native fold have rapid fibril formation (37 °C, shaking at 750 rpm, 0.5 mM SDS). Low occurred as resolved by cryo-EM. During restructuring, the internal concentrations of SDS are typically used as a destabilizing agent to 6,24 disulfide bond is retained and hydrophobic core residues become accelerate fibril formation of amyloid proteins .Aliquotswere 14,15 surface-exposed . Like in other amyloid diseases, the growth taken at different timepoints during incubation and separated into phase of AL fibrilsisprecededbyacomparably long and rate- soluble and insoluble fractions by centrifugation. Analysis of the 16–18 limiting lag phase in vitro . So far, little is known about the fractions by SDS-PAGE and quantification of the bands showed that structural events taking place during this phase. For different amyloid during the first hour of incubation, the amount of soluble V domain diseases, oligomeric intermediates of the amyloidogenic proteins and did not change significantly (Fig. 1a). Then, a decrease of soluble Pat- 16,19–22 a nucleated polymerization mechanism have been suggested . 1 and a concomitant increase of Pat-1 in the insoluble fraction were Capturing these species remains challenging, due to their transient observed over time. The half times of the reactions were very similar, appearance in low concentrations and the high energy states of the with a t of 1.79 h ± 0.11 for the decrease of the soluble form and t 1/2 1/2 16,18 specificintermediates . In AL amyloidosis, the presence of oli- of 1.71 h ± 0.30 h for the increase in the insoluble species (Fig. 1b). gomers forming during the lag phase and the accompanied structural After 2 h, almost no protein was left in the soluble fraction. Fibril transitions prior to amyloid formation have not been investigated so formation was confirmed by the measurement of the ThT fluores- far. Understanding the pathway and molecular mechanism of reac- cence. Fibrils started to form after 2 h and the half time of the tions preceding fibril formation of pathogenic LCs is important to reaction was 3.5 h (Fig. 1b).After 4h,nofurther increase in ThT identify potential therapeutic intervention points at early stages of the fluorescence was visible, thus no further fibril growth seems to take disease. Here, we elucidated the processes taking place during the lag place while ongoing structural rearrangement cannot be excluded. phase prior to fibril formation of the well-studied pathogenic V The presence of fibrils was confirmed by TEM micrographs (Fig. 1c). domain Pat-1 .Weidentified intermediate oligomeric species on the For Pat-1 L15P L82Q, fibril formation occurred even faster while the fibril pathway and associated structural rearrangements using a broad germline did not form fibrilsasconfirmed by ThT fluorescence and range of biophysical analyses. TEM (Supplementary Figs. 1, 4d, h). When we transferred the Pat-1 at different timepoints during the lag phase to lower temperatures (4 °C or 20 °C) and stopped shaking, ThT fluorescence analysis Results showed that no further fibril formation occured over a time period of Fibril formation is preceded by the disappearance of soluble V 3 h (Supplementary Fig. 2). This allowed us to dissect the lag phase monomers.In AL amyloidosis, fibril formation in vitro is usually and perform detailed analyses of the samples. preceded by a long lag phase after which fibrils form in a rapid and Fig. 1 Transition from soluble V Pat-1 to insoluble fibrils. a Soluble and insoluble fractions of Pat-1 V during incubation at amyloid-promoting conditions. L L Samples were taken at the timepoints indicated and run on an 18% SDS-gel. b Quantified and normalized band intensities of the SDS-gel are shown in turquoise for the soluble V domain (t = 1.79 h) and in black for the insoluble fraction (t = 1.71 h). Shades represent the SEM of n = 2. Fibril formation L 1/2 1/2 of Pat-1 V as monitored by ThT fluorescence is shown in pink (t = 3.48 h). Values were fit to a Boltzmann function Shades represent the SEM of n = 3. L 1/2 c TEM micrograph of fibrils formed after 3 h of incubation. The scale bar represents 200 nm. 2 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE Fig. 2 Oligomeric species present in the lag phase of the amyloid pathway of Pat-1 V . AUC sedimentation profiles over time of a monomeric and dimeric species and b oligomeric species. c Normalized AUC peak quantification of monomeric (yellow) and oligomeric (purple) species during the lag phase of fibril formation. The data for monomeric species were fit to a Boltzmann function, the data for oligomeric species were fit to a Gaussian function. Shade of the ThT Fluorescence represents the SEM of n = 3. d Section of TEM micrographs shown in Fig. S5 of the oligomer and fibril formation over time. Scale bar represents 50 nm. Different oligomeric intermediates are part of the pathway to hour of incubation. After that, they cannot be detected unam- fibril formation. Due to the denaturing conditions, SDS-PAGE biguously due to the lower concentration of monomers and the analysis does not provide information on noncovalent structural lack of resolution provided by the fitting software. TEM micro- alterations, specifically oligomerization, taking place during the graphs confirmed the presence of oligomeric species after 15 min lag phase. Thus, to determine potential changes in quaternary incubation of Pat-1 and immediately before fibril formation after structure, we performed analytical ultracentrifugation (AUC). For 1.75 h in Pat-1 L15P L82Q. Over time, the oligomer amount a better resolution of the different species we increased the pro- increases and fibrils for Pat-1 could be detected after 3 h and after tein concentration. The sedimentation analysis revealed that at 2 h for Pat-1 L15P L82Q consistent with the results of the ThT the start point (0 h), Pat-1 was present mainly as a monomer with assay (Fig. 2d, Supplementary Figs. 1, 4d, 5). Large oligomers a small fraction of dimers (Fig. 2a, Supplementary Fig. 3). After were visible in the TEM micrographs in a lower number, which 15 min incubation, a range of oligomers became visible as four could be due to their dissociation under the acidic conditions distinct peaks in the range from 4 to 20 S in addition (Fig. 2b, during negative staining. The germline formed oligomeric species Supplementary Fig. 3). The amount of oligomeric species with a clustering morphology. However, here no fibrils but increased during the first 30 min, while the monomer fraction amorphous aggregates were formed (Supplementary Fig. 4e-h). decreased. After 45 min, the oligomeric species reached a peak, while about half of the sample was still monomeric. After that The V secondary structure changes during oligomerization. timepoint, both oligomers and monomers decreased and dis- As the native V domain and AL fibrils differ substantially in side appeared completely, and fibrils evolved (Fig. 2a–c). At each 14,15 chain interactions , conformational remodeling of V has to timepoint tested, the concentration of oligomers was lower than occur in the lag phase. To detect potential secondary structure the concentration of monomers, as determined by the area under changes, we followed the reaction over time by far UV (FUV) the curves (Fig. 2a–c). For Pat-1 L15P L82Q, the appearance of circular dichroism (CD) spectroscopy. At the start of the reaction, oligomers before fibril formation was also observed (Supple- Pat-1 V showed a minimum at 218 nm followed by an amplitude mentary Fig. 4a, b). Since the monomer peak (1.5 S) shows a L as expected for the antiparallel β-sheet native structure of Ig shoulder towards higher S values (2.1 S) from the beginning of the domains (Fig. 3a) . After 0.5 h incubation, the amplitude of the incubation onwards, we conclude that initially V dimers form in minimum at 218 nm increased. The alterations observed during Pat-1 and Pat-1 L15P L82Q (Fig. 2a, Supplementary Figs. 2, 4a-c). the lag phase indicate a reorientation and partial unfolding of These dimers remain present in low amounts during the first NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 Fig. 3 Secondary structure changes during oligomer formation monitored by FUV CD-and FTIR spectroscopy. a Comparison of changes in the secondary structure of Pat-1 in the lag phase. The spectra show the average of 10 individual scans. b Secondary structure analysis of the CD spectra of Pat-1 using the BeStSel algorithm. c Secondary structure changes in the lag phase monitored by ATR-FTIR after H/D exchange. The spectra of the amide I and II region were scaled to an absorbance of 1 and baseline corrected. d Comparison of the shift of the minimum in the spectra of different timepoints during the lag phase (blue), the change of intensity at 210 nm (green) with the fibril formation kinetics followed by ThT fluorescence (pink). Shades of the ThT fluorescence represent the SEM of n = 3. secondary structure elements. This trend continued further dur- domain rearrangements during oligomer formation and rear- ing 1.5 h of incubation. After 1.75 h, the CD minimum shifted to rangement. Upon fibril formation, the signal remains constant. 220 nm and the amplitude increased. At the same time, a max- Taken together, the CD analysis suggests that despite the imum at 204 nm emerged. The shift of the minimum and the apparent rearrangement of secondary structure, secondary simultaneous intersection of the x-axis at higher wavelengths structure elements remain present during the lag phase. Con- indicates the formation of parallel β-sheets . Thus, this time- comitant with fibril formation, the CD data suggest a structural point marks an important conformational rearrangement. During rearrangement event leading from antiparallel to cross β-sheets. the lag phase, the high voltage (HT) at the detector, which reflects To further analyze secondary structure changes during the lag the absorption of light by the sample, also changed. The signal phase we used attenuated total reflection Fourier transform decreased during the lag phase, which means more light reaches infrared spectroscopy (ATR-FTIR). In order to minimize the detector (Fig. S6). This in turn suggests that structural contributions in the Amid I region originating from residual changes affecting the signal take place. After 2 h, the minimum water bound to buffer components we recorded spectra after H/D shifted further to 224 nm and the maximum at 204 nm was more exchange (Supplementary Fig. 7c). The overall content of β- pronounced. This maximum is indicative of supramolecular β- structure remains unchanged in samples of mono- and oligomers sheet-rich amyloid structures . A similar effect was also observed while fibril formation results in a peak shift from 1638 to 1625 cm −1 for Pat-1 L15P L82Q but not for the germline WT-1 (Supple- because of intermolecular β-sheet stacking in the fibrillar mentary Fig. 7a). Analysis of the CD data with the BeStSel structure (Fig. 3c) . Concomitantly, the shoulder around −1 algorithm revealed an decrease of antiparallel β-strands with a 1685 cm disappears. This observation indicates a decrease of 30 −1 slight increase in parallel β-strands and the unfolded fraction for antiparallel β-sheet content . The peaks at 1590 and 1515 cm Pat-1 and Pat-1 L15P L82Q but not for WT-1 (Fig. 3b, Supple- mainly belong to side chain absorptions and Tyr, respectively . mentary Fig. 7b). The shift of the minimum in the FUV CD spectra over time precedes the appearance of fibrils, as the ThT fluorescence starts to increase after the wavelength shift of the CD The V domain structure changes during oligomerization.To minimum to 224 nm (Fig. 3d). Interestingly, even before the shift monitor structural changes in the core of the V domain, we in the CD minimum occurs, the signal of the amplitude at lower analyzed the fluorescence of the single intrinsic tryptophan as a wavelengths decreases in the first hour, which might indicate sensitive and specific probe since its emission intensity is quen- partial unfolding. The changes are particularly noticeable at a ched in the native state by the disulfide bridge located in the core wavelength of 210 nm, which is why these were used for com- of the β-barrel and becomes higher upon unfolding . Thus, parison in Fig. 3d. The following increase of the amplitude conformational changes involving the protein core can be mon- between 1 and 2 h of the lag phase seems to correspond to itored by a change in the fluorescence amplitude of tryptophan. 4 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE Fig. 4 Structural changes of Pat-1 during the lag phase. a Change of tryptophan fluorescence. Each emission spectrum is the average of three individual scans. b Change of surface hydrophobicity followed by ANS fluorescence. c Change of tryptophan fluorescence upon quenching with 0–1 M acrylamide. The curves show the respective F /F values of the Stern-Volmer equation, Error bars represent the SEM of n = 3. d Comparison of the normalized change of ANS fluorescence (gray), Trp exposure (orange) and the quenching constant K of the acrylamide quenching calculated by Eq. (1) (purple) in comparison to changes in the ThT signal (pink). Shades represent the SEM of n = 3. When we analyzed changes in the fluorescence emission during detected by CD spectroscopy (Figs. 3a, 4b; Supplementary the lag phase, we observed a steady increase from the beginning Fig. 8c). onwards indicating that structural rearrangements take place in To extend our analyses of conformational rearrangements via the protein core (Fig. 4a). The fluorescence intensity reached its the accessibility of tryptophan, fluorescence quenching by maximum after 3 h, at the same time when fibrils have formed acrylamide was assessed. The quenching is stronger the more (Fig. 4d). Pat-1 L15P L82Q and WT-1 also showed an increase in solvent accessible the tryptophan residue is. There is a steep tryptophan fluorescence which occurred rapidly upon fibril for- increase in fluorescence quenching already at the beginning of the mation for the double mutant while the germline shows no fibril lag phase and the amplitude continues to grow up to 1 h. This formation (Supplementary Figs. 4, 8a, b). indicates that the structural rearrangement we observed via To further probe structural changes of the domain surface, we changes in tryptophan fluorescence involves the rapid reposition- used the fluorophore ANS which binds specifically to surface- ing of the buried tryptophan to a solvent-exposed position exposed hydrophobic patches in a protein. Upon binding, the (Fig. 4c). fluorescence intensity increases and the emission maximum shifts To obtain further structural insight in the changes of the to lower wavelengths . Following ANS fluorescence during the tryptophan environment, we employed red edge excitation shift lag phase, we observed an increase in ANS fluorescence intensity (REES) spectroscopy . The REES effect is driven by the dipole from the beginning of the incubation until a maximum was interactions of the fluorophore with its surrounding: a rigid or a reached after 2 h. The increase in surface hydrophobicity occurs completely solvent-exposed surrounding leads to a smaller effect for both, Pat-1 L15P L82Q and the germline, although to a much than a flexible protein present in different conformational greater extent in the double mutant (Supplementary Figs. 8c, d). states . For the Pat-1 V domain, we observed an increase of Since the germline also forms oligomers and aggregates, an the center of spectral mass (CSM) after 0.25 h compared to the increase in ANS binding is reasonable. Thus, our results indicate start of the reaction (Fig. 5a, b). This implies a strong increase in that rearrangements in the environment of the tryptophan in the solvent exposure of the tryptophan at the beginning of the lag core also leads to changes in the surface hydrophobicity in Pat-1 phase due to a domain opening. After 0.25 h, a second, slower and its double mutant. This domain opening and increasing phase became apparent, which increased over time (Fig. 5B). The solvent exposure of the inner core of the VL domains occurs first higher REES effect suggests a rugged free energy landscape of the in the lag phase, followed by changes in secondary structure as V domain at the beginning of the reaction, which decreases NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 Fig. 5 The REES effect of Pat-1 V during the lag phase. a Change in the CSM for the corresponding excitation wavelengths from 280 to 300 nm. b Conclusion of the REES effect (orange) and solvent exposure (black) at the different timepoints. Measurements were carried out in triplicates. Error bars respresent the respective SEM of the triplicates. quickly, indicating that there are less conformational states phase, a specific pattern of higher oligomers was detected which available, which influence tryptophan fluorescence relaxation. rearranged before the assembly into fibrils starts. Concomitantly, The smaller REES effect is not due to an extensive unfolding of the monomers decrease during the lag phase. The oligomer the protein, as in the presence of 6 M urea a pronounced shift to fraction shows a complex behavior that slightly differs between smaller values for the REES effect was observed (Supplementary AUC and TEM analysis. According to AUC analysis, the amount Fig. 9c). Pat-1 L15P L82Q shows a similar effect concerning of oligomers first increases and then decreases until they com- increased solvent exposure and decreasing REES effect (Supple- pletely disappear when fibrils are formed. This hints at the for- mentary Fig. 9c). The later stays the same for the germline while mation of a nucleus prior to fibril formation as proposed for other 43,44 also an increase in the solvent exposure can be observed matching amyloid reactions . The cooperative transition to the fibrillary the fluorescence data (Supplementary Figs. 8b, 9b). Thus, the two state supports the view that a specific conformational state of the phases of the REES kinetics during the lag phase may be related to oligomers is the seed for polymerization. Accordingly, the oli- the initial partial unfolding with a concomitant higher solvent gomeric species were always present in lower amounts compared accessibility or a higher rigidity due to dimerization followed by to soluble monomer/dimer fraction or insoluble fibrils as oligomer formation and rearrangement in the oligomer. In described for other amyloids . TEM micrographs of Pat-1 contrast to the fibril forming proteins, the germline accesses a showed oligomers that could correspond to hexamers, which lower amount of conformational states as indicated by the overall increase during the lag phase in number but not in size; the Pat-1 lower REES effect (Supplementary Fig. 9c) double mutant L15P L82Q formed similar oligomers rapidly before fibril formation. As negative staining involves the incu- bation in a low pH uranyl acetate solution, we assume that this Discussion dissolves the higher oligomers. Thus, the hexamer is the most The conformational switch from the native to the fibrillary state stable oligomeric species. In contrast to Pat-1, the WT-1 oligomer in amyloidosis comprises a lag phase, including a primary species exhibited clustering and unspecific aggregation in TEM nucleation step, a transition phase in which fibrils start to form analysis. and elongate, and a final plateau phase in which fibrils are present 16,35 According to the known secondary and tertiary structure of the and an equilibrium is reached . The same general scheme was 6,36–38 monomers as well as the fibrils, major conformational rearran- assumed to apply for AL amyloidosis . However, the gements have to occur in V prior to fibril formation as the molecular events occurring in early phases of the fibrillation domain consists of a two-layer sandwich structure composed of process were largely unknown. In this study, we determined the antiparallel β-strands, whereas the amyloid fibrils in AL amyloi- conformational transitions and the molecular species formed in dosis exhibit a cross-β sheet topology consisting of parallel β- the lag phase that predispose the pathogenic V domain to fibril 14,15,45,46 sheets . The single tryptophan residue buried in the core formation. of the domain is an excellent spectroscopic probe for con- We show that, starting from the monomeric patient LC trun- 6 formational changes as it reports on variations in its local cation Pat-1 , small amounts of dimers are formed in the first environment . Analysis of the intrinsic tryptophan fluorescence hour of the lag phase. They might be caused by a dynamic revealed that the microenvironment of the tryptophan residue in equilibrium between these two states. Later on, the peak shoulder the core starts to change already in the very beginning of the lag is not readily visible, however, we presume that a fraction of phase as seen by an increase of overall fluorescence intensity. This dimers is still present. As the dimeric fraction is the first obser- effect increases with time indicating that the domain structure vable step that differs between native and fibril-inducing condi- increasingly changes. The distancing of the tryptophan residue tions, these dimers reflect conformational changes which lead to 6 from the disulfide bond and the enhanced accessibility reflects the destabilized monomers which finally form fibrils . Of note, we opening of the β-barrel structure while the overall secondary consider these dimers as non-native species induced by the structure elements still remain largely intact as judged from the increasing hydrophobic surface at a beginning unfolding. They FUV CD measurements. This domain opening goes along with need to be differentiated from the native dimers observed for a the formation of small amounts of dimers as seen in the AUC number of LCs. In this context, the dimers either do not interfere 36,39 sedimentation profiles. Thus, we hypothesize that even though with fibril formation or they may exert a protective role 40–42 the overall secondary structure of the V monomer does not against fibrillization . In addition, after 0.25 h of the lag 6 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE change during initial oligomerization, a state appears in which the Compared to the disease-related V domains, the germline packing starts to rearrange, and hydrophobic residues become behaves differently in important aspects. While it also undergoes surface-exposed. ANS binding experiments show an early partial unfolding that leads to an increase in the tryptophan increase of surface-exposed hydrophobicity, which seems to fluorescence and a slight increase in surface hydrophobicity, the coincide with domain opening and oligomer formation. At this conformational state of the germline remains constant during the step, a further conformational reorganization occurs that fosters lag phase and there are no observable changes in the secondary oligomerization via hydrophobic interactions, as observed for structure. The oligomeric species formed end in amorphous 47–49 other amyloidogenic proteins . The acrylamide quenching aggregates. experiments give additional support to the idea of conformational In summary, our biophysical analyses reveal a multi-step rearrangements during oligomer formation. It monitors the conformational transition from a folded monomeric β-barrel accessibility of a quencher to a fluorophore and thus is a marker domain into an amyloid fibril (Fig. 6). It starts with an initial for solvent accessibility. The increase of the quenching constant K domain opening and partial unfolding in which the β-strands of from the beginning of the lag phase onwards implies a higher the V monomers are preserved. The increase in surface hydro- solvent exposure of the buried tryptophan along with the partial phobicity fosters dimerization and concomitant assembly into unfolding of the V domain. REES experiments further revealed hexamers and multiples of hexamers. In this process further that an increased solvent accessibility of the tryptophan residue, rearrangements occur, which reduce conformational flexibility. monitored by the CSM at an excitation wavelength of 280 nm, Consistent with the idea that the oligomers are the species with takes place. This again supports the notion of domain opening the highest Gibbs free energy, they are present only transiently concluded from changes in tryptophan fluorescence and ANS and at low concentration. It was previously reported that the life binding. The increase of CSM values at the y-intercept goes along time of oligomers of amyloid proteins can vary significantly. Also, with a decrease in the REES effect. The emission spectrum of a oligomers may have a higher tendency for dissociation to fluorophore is highly dependent on its environment since it is monomers than for fibril formation . influenced by the dipole interaction with water. A more solvent- In the context of our study, it is useful to compare the concept exposed surrounding thereby leads to a smaller REES effect, as established for AL amyloidosis with other fibril forming proteins. shown by the unfolding of Pat-1 with urea. The magnitude of the The dialysis-related amyloid precursor β -microglobulin also REES effect can provide information about the free energy consists of an immunoglobulin fold. It was shown that starting landscape of a protein . A higher REES effect is based on a from the natively folded monomer, an unfolded state emerges 39,52–54 higher number of discrete conformational states as seen for Pat-1 and subsequent oligomerization precedes fibril formation . and Pat-1 L15P L82Q. Upon oligomerization this REES effect Clustering following unfolding has also been suggested for amy- decreases, which also implies a decrease in conformational states. loidosis involving α-helical precursors . Our study shows that In contrast, WT-1 shows unfolding events but the accessible the decisive committing step for the amyloid pathway occurs in conformational states do not increase. Since the fluorescence the context of the oligomer. In contrast, unfolding and initial intensity does not decrease during the lag phase and during fibril oligomerization were also observed for the nonamyloidogenic formation, the tryptophan does not shift back to its position in germline protein. However, this protein subsequently ends up in close proximity of the quenching disulfide bond. These results are amorphous aggregates. Thus, it seems that only once the oligo- in excellent agreement with the cryo-EM structure of AL fibrils, mers undergo a critical structural rearrangement nuclei are where the conserved tryptophan residue is found close to the formed, which are rapidly transformed into amyloid fibrils. For fibril surface. Furthermore, the disulfide bond is intact but the further insights into the structure of the oligomers, high resolu- interactions of residues including the tryptophan differ sig- tion methods like cryo-EM or solid state NMR need to be applied 14,50 nificantly from that in the native protein . We assume that in the future. Especially, since pathogenic effects could be ascri- after initial unfolding events and partial dimer formation from bed not only to insoluble fibrils but also to oligomers . Fur- folded monomers, a subsequent fast association of monomeric or thermore, a molecular understanding of the conformational dimeric species to oligomers occurs. Preceding fibril formation transitions in the lag phase of different amyloidoses may result in but after the initial unfolding events, a shift in the FUV CD the emergence of general concepts across diseases. minima became visible which we assume represents the rear- rangement of the β-sheets. The change of local minima fits to the Methods switch of antiparallel to parallel β-sheet fold . Thus, the CD All chemicals were from Merck (Darmstadt, Germany) or Sigma (St. Louis, USA). results indicate that this timepoint marks further important All measurements were carried out in PBS buffer containing small amounts of structural rearrangements induced by intermolecular interactions. 24,57,58 SDS (10 mM Na HPO ×2 H O; 1.8 mM KH PO ; 2.7 mM KCl; 137 mM 2 4 2 2 4 The rapid reaction into fibrillary structures with features of a NaCl; 0.5 mM SDS) at pH 7.4 and 37 °C, unless otherwise stated. Data were supramolecular β-sheet formation reveals that the rearranged analyzed using Origin 2019. species is transient, meta-stable and thus potentially presents the nucleus for fibril formation. The β-sheet formation in fibrils was Expression and purification of Pat-1, Pat-1 L15P L82Q, and WT-1. The origin of also observed in FTIR, the rapid switch after ~2 h could not be the Pat-1 sequence and the recombinant expression and purification of Pat-1, Pat-1 resolved in detail. However a difference in the amide II region L15P L82Q, and WT-1 was described before. Also, the generation of the point emerged. Of note, the FTIR measurements required extensive mutations with the pirmers TAGCGGTAGCCCGGGTCAGAGCATTA (+) and ACGCTTGCAGGCTGGGTC (−) has been previously described. In brief, the sample processing and could not be conducted in a time-resolved plasmids were transformed in E. coli BL21 (DE3)-star cells and protein expression manner similar to the CD. In this context, it should be noticed took place at 37 °C overnight. Cells were harvested and inclusion bodies were that in general conditions were kept identical between the dif- prepared. The pellet was solubilized and unfolded in 25 mM Tris-HCl (pH 8), ferent methods used. However, additives like ANS or ThT, or 5 mM EDTA, 8 M urea, and 2 mM β-mercaptoethanol at room temperature for a slightly different conditions like in the AUC experiments might minimum of 2 h. Afterwards, the protein was loaded onto a Q-Sepharose anion exchange column equilibrated in 25 mM Tris-HCl (pH 8.0), 5 mM EDTA and 5 M potentially impact the conformational equilibrium. However, the urea. The LCs and V s were eluted in the flow-through fractions and refolded by methodologies have been well established for the addressed dialysis against 250 mM Tris-HCl (pH 8.0), 100 mM L-Arg, 5 mM EDTA, 1 mM questions and the results obtained are in line with each other, oxidized glutathione and 0.5 mM reduced glutathione at 4 °C overnight. To remove supporting the individual evidence. aggregates and impurities, the refolded proteins were purified using a Superdex 75 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 Monomers/Dimers Oligomers Fibrils rearrangement Time Folded Monomer Oligomer Oligomer from Fibrils monomer with partially with partially structually rearranged opened structure opened monomers monomers - nucleus Fig. 6 Model of the events taking place in the lag phase of fibril formation. Folded V monomers undergo a partial structural opening whereupon they assemble to dimers, and oligomers (hexamers and multiples of hexamers). Prior to fibril formation, the monomers in these oligomers structurally rearrange resulting in parallel β-sheets. This variant is the nucleus of fibrillization. The above bars show the concentration of each species at the respective timepoint of the fibril formation process. The concentration of oligomers is low at all timepoints. 16/60 gel-filtration column (GE Healthcare, Uppsala, Sweden) equilibrated in PBS 70 from Bruker, Germany). The samples were equilibrated in the measurement buffer. Recovery and purity of intact proteins were analyzed by SDS-PAGE. chamber under nitrogen flow for 30 min and subsequently incubated with D O saturated nitrogen flow for H/D exchange of bound water fractions. The kinetics of −1 H/D exchange was repeatedly recorded with 32 scans at 2 cm resolution until the Sample preparation for lag phase analysis. For studying the different time- spectra remained unchanged. After 50 min the final spectra were recorded, accu- points, for all measurements unless otherwise stated, 15 µM of Pat-1 were prepared mulating 1024 scans. The linear baseline was subtracted from spectra for further in the prior described assay buffer in a 1.5 mL Eppendorf reaction tube and were analyses and documentation. incubated at 37 °C ± 2 and 750 rpm shaking in a Thermomixer compact (Eppen- dorf, Hamburg, Germany). For 0 h timepoints, the sample was put on 37 °C and removed before shaking. All other timepoints were removed after the corre- 8-Anilino-1-naphtalenesulfonic acid (ANS) and Tryptophan Fluorescence. For measurement of ANS binding, samples were taken at different timepoints and sponding time shaking. All samples were kept on ice afterwards. To confirm the interruption of fibril formation at the different timepoints the shaking was stopped incubated with 150 µM ANS for 1 h. Spectra were recorded from 400 to 650 nm with an excitation of 380 nm. For intrinsic Tryptophan fluorescence measurements and the samples were kept on ice or at 25 °C for 3 h prior to ThT fluorescence analysis. spectra were recorded from 300–450 nm with an excitation wavelength of 280 nm. All fluorescence measurements were carried out using a Jasco FP-8500 Spectro- fluorometer (JASCO, Pfungstadt, Germany) at 25 °C. The settings included exci- Analytical ultracentrifugation (AUC). AUC measurements were carried out tation and emission bandwidth of 5 nm each, 4 s response time, a data interval of using an Optima AUC (Beckman, Krefeld, Germany) equipped with absorbance 1 nm and 200 nm/min scan time.. For Pat-1 L15P L82Q and WT-1 spectra were optics. The protein concentration for the measurements was 30 µM due to a low recorded at a Tecan Infinite 200 PRO M Nano with a data interval of 1 nm and an data resolution at lower concentrations. A total volume of 350 µL per sample was amplification of 100. Depicted spectra show the average of three individual mea- loaded into assembled cells with quartz windows and 12 mm-path-length charcoal- surements. Samples were taken from a 15 µM protein solution incubated at 37 °C filled epon double-sector centerpieces. The measurements were performed at and 700 rpm. 42,000 rpm in an eight-hole Beckman-Coulter AN50-ti rotor at 20 °C. Sedi- mentation was continuously scanned with a radial resolution of 10 µm and mon- Acrylamide quenching. For Acrylamide quenching, samples were taken at dif- itored at 280 nm. Data analysis was carried out with software SEDFIT using the 59,60 continuous c(S) distribution mode . ferent timepoints and 0 M to 1 M acrylamide in 1 × PBS in 0.2 M steps were added to 5 µM of the protein. The tryptophan fluorescence was recorded from 300 to 400 nm with an excitation of 280 nm. The measurement was carried out at 37 °C at Far-UV (FUV) circular dichroism (CD) measurements. For Pat-1, FUV CD a Tecan Infinite 200 PRO M Nano with an amplification of 157 and a data spectra were recorded from 197–260 nm using a Chirascan-plus CD spectrometer interval of 1 nm. The Stern-Volmer quotient F /F was calculated with the maximal (Applied Photophysics, Leatherhead, England). Measurements were recorded with fluorescence intensity at 332 nm, while F is the value at 0 M and F the fluorescence a bandwidth of 1.0 nm in 1.0 nm steps and 0.5 s time per point at a temperature of 61,62 intensity at the respective acrylamide concentration . The raw data was linearly 37 °C. All measurements were performed using a 15 µM protein solution in a fitted. The slope represents the quenching constant regarding Eq. (1): quartz cuvette with 1 mm pathlength. The spectra show an average of 10 individual measurements. For Pat-1 L15P L82Q and WT-1 FUV CD spectra were recorded from 197 to ð1Þ 260 nm using a Jasco J-1500 (JASCO, Pfungstadt, Germany). Measurements were ¼ 1 þ KQjj recorded with a data pitch of 0.1 nm and a scanning rate of 20 nm/min time per point at a temperature of 37 °C. All measurements were performed using a 15 µM protein solution in a quartz cuvette with 0.5 mm pathlength. The spectra show an average of 20 individual measurements. Red edge excitation shift (REES). The REES effect at different timepoints was conducted at a Jasco FP-8500 Spectrofluorometer at 25 °C with a scanning rate of Attenuated total reflection Fourier transform infrared spectroscopy (ATR- 500 nm/min. The setup included a bandwidth of 5 nm for excitation and emission, FTIR). After the respective timepoint of the lag phase, samples were dialyzed in a response time of 2 s and an interval of 1 nm. The tryptophan emission was 10 mM Na HPO ×2 H O; 1.8 mM KH PO overnight and 250 µl were dried on monitored from 315 to 400 nm with an excitation scan from 280 to 300 nm. For 2 4 2 2 4 one side of a germanium crystal under nitrogen flow. The crystal was mounted into data analysis, the center of spectral mass (CSM) was calculated as reported a home-made, gas-tight holder and the latter in the FTIR spectrometer (VERTEX previously . 8 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications Fibril Formation Conc. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE Thioflavin T (ThT) assay. Fast fibril formation assays followed by ThT fluores- 10. Glenner, G. G., Harbaugh, J., Ohms, J. I., Harada, M. & Cuatrecasas, P. An cence were performed in 1.5 mL Eppendorf tubes. For Pat-1, kinetics were followed amyloid protein: the amino-terminal variable fragment of an immunoglobulin by measuring at different timepoints in a 10 × 2 mm quartz cuvette using a light chain. Biochem. Biophys. Res. Commun. 41, 1287–1289 (1970). FluoroMax-4 spectrofluorometer (Horiba Jobin Yvon, Bensheim, Germany) and a 11. Stevens, F. J. et al. A molecular model for self-assembly of amyloid fibrils: Jasco FP-8500 Fluorescence Spectrometer (JASCO, Pfungstadt, Germany). Kinetics immunoglobulin light chains. Biochemistry 34, 10697–10702 (1995). were followed by measuring at 440 nm excitation and 480 nm emission with slit 12. Merlini, G. & Bellotti, V. Molecular mechanisms of amyloidosis. N. Engl. J. widths of 3 nm or 2.5 nm for excitation and 4 nm or 5 nm for emission, respec- Med. 349, 583–596 (2003). tively. For Pat-1 L15P L82Q and WT-1, kinetics were followed by measuring at 13. Pepys, M. B. Amyloidosis. Annu. Rev. Med. 57, 223–241 (2006). different timepoints in a 96-well UV star plate (Greiner Bio-One, Frickenhausen, 14. Radamaker, L. et al. Cryo-EM structure of a light chain-derived amyloid fibril Germany) at a Tecan Infinite 200 PRO M Nano with an amplification of 60 and a from a patient with systemic AL amyloidosis. Nat. Commun. 10,1–8 (2019). data interval of 1 nm, The ThT fluorescence was measured from 460 to 500 nm 15. Swuec, P. et al. Cryo-EM structure of cardiac amyloid fibrils from an with an excitation wavelength of 440 nm. The data were plotted by using the ThT immunoglobulin light chain AL amyloidosis patient. Nat. Commun. 10,1–9 fluorescence at 480 nm. To remove oligomeric species and aggregated protein and (2019). prevent seed formation, monomer isolation was performed prior to the experiment 16. Arosio, P., Knowles, T. P. J. & Linse, S. On the lag phase in amyloid fibril TM by ultracentrifugation in an Optima MAX-E ultracentrifuge, (Beckman, Krefeld, formation. Phys. Chem. Chem. Phys. 17, 7607–7618 (2015). Germany). 17. Harper, J. D. & Lansbury, P. T. Models of amyloid seeding in Alzheimer’s disease and scrapie: mechanistic truths and physiological consequences of the Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). SDS time-dependent solubility of amyloid proteins. Annu. Rev. Biochem. 66, Gels were cast in two steps pouring first the 18% separating gel solution and then 385–407 (1997). the 5% stacking gel solution. For sample preparation, samples taken from 15 µM 18. Monsellier, E. & Chiti, F. Prevention of amyloid-like aggregation as a driving Pat-1 solution at different timepoints were centrifuged and separated into super- force of protein evolution. EMBO Rep. 8, 737–742 (2007). natant and pellet fraction. The pellet was washed twice with PBS. Supernatant 19. Lesné, S. et al. A specific amyloid-β protein assembly in the brain impairs samples were diluted with 5 × reducing Laemmli buffer, pellets were dissolved in memory. Nature 440, 352–357 (2006). 1 × reducing Laemmli buffer. Both were heated at 95 °C for 5 min for denaturation. 20. Kastritis, E. & Dimopoulos, M. A. Recent advances in the management of AL 20 µL of the respective samples were applied into the gel pockets. The gel was run Amyloidosis. Br. J. Haematol. 172, 170–186 (2016). at 30 mA for 40 min and stained with Fairbanks A (25% Isopropanol, 10% Acetic 21. Kayed, R. et al. Common structure of soluble amyloid oligomers implies Acid, 0.05% Coomassie R) and destained with Fairbanks D (10% Acetic Acid) common mechanism of pathogenesis. Science 300, 486–489 (2003). afterwards. 22. Sikkink, L. A. & Ramirez-Alvarado, M. Cytotoxicity of amyloidogenic immunoglobulin light chains in cell culture. Cell Death Dis. 1, 1 (2010). Transmission electron microscopy (TEM). TEM samples were prepared by 23. Morgan, G. J. & Kelly, J. W. The kinetic stability of a full-length antibody light pipetting 10 µL samples onto a 200-mesh activated copper grid and incubating for chain dimer determines whether endoproteolysis can release amyloidogenic 1 min. The samples were washed with 2 × 10 µL H O and negatively stained with variable domains. J. Mol. Biol. 428, 4280–4297 (2016). 8 µl of a 1.5% (w/v) uranyl acetate solution for 1 min. Excess solutions were 24. Yamamoto, S. et al. Low concentrations of sodium dodecyl sulfate induce the removed with a filter paper. TEM micrographs were recorded on a JEOL JEM-1400 extension of β2-microglobulin-related amyloid fibrils at a neutral pH. Plus transmission electron microscope (JEOL Germany GmbH, Freising, Ger- Biochemistry 43, 11075–11082 (2004). many) at 120 kV. 25. Goto, Y. & Hamaguchi, K. Light role of the intrachain of the constant disulfide fragment bond in the conformation and stability of the immunoglobulin by measuring Intact CL was more resistant than reduced CL to guanidine In a Statistics and reproducibility. All representative TEM micrographs in this pub- previous paper (1) we reported on the refolding of B. J. Biochem. 86, lication have been chosen from 3 to 20 images taken from a negative stained 200- 1433–1441 (1979). mesh activated copper grid. 26. Bolotina, I. A., Chekhov, V. O. & Lugauskas, V. Y. Determination of the secondary structures of proteins by circular dichroism spectra. Calculation of Reporting summary. Further information on research design is available in the Nature the protein basic circular dichroism spectra for antiparallel and parallel β‐ Research Reporting Summary linked to this article. structures and β‐bends. Int. J. Quantum Chem. 16, 819–824 (1979). 27. Sahoo, B. R. et al. A cationic polymethacrylate-copolymer acts as an agonist for β-amyloid and an antagonist for amylin fibrillation. Chem. Sci. 10, Data availability 3976–3986 (2019). All data generated during this study are available in the article and supplementary 28. Micsonai, A. et al. BeStSel: a web server for accurate protein secondary information or from the corresponding author. Source data are provided with this paper. structure prediction and fold recognition from the circular dichroism spectra. Nucleic Acids Res. 46, W315–W322 (2018). Received: 18 December 2020; Accepted: 25 October 2021; 29. Zandomeneghi, G., Krebs, M. R. H., McCammon, M. G. & Fändrich, M. FTIR reveals structural differences between native beta-sheet proteins and amyloid fibrils. Protein Sci. 13, 3314–3321 (2004). 30. Goormaghtigh, E., Ruysschaert, J. M. & Raussens, V. Evaluation of the information content in infrared spectra for protein secondary structure References determination. Biophys. J. 90, 2946–2957 (2006). 1. Quock, T. P., Yan, T., Chang, E., Guthrie, S. & Broder, M. S. Epidemiology of 31. De Meutter, J. & Goormaghtigh, E. Evaluation of protein secondary structure AL amyloidosis: a real-world study using US claims data. Blood Adv. 2, from FTIR spectra improved after partial deuteration. Eur. Biophys. J. 50, 1046–1053 (2018). 613–628 (2021). 2. Baden, E. M., Sikkink, L. A. & Ramirez-Alvarado, M. Light chain amyloidosis 32. Horowitz, P. M. & Criscimagna, N. L. Differential binding of the fluorescent —-current findings and future prospects. Curr. Protein Pept. Sci. 10, 500–508 probe 8-anilinonaphthalene-2-sulfonic acid to rhodanese catalytic (2009). intermediates. Biochemistry 24, 2587–2593 (1985). 3. Gertz, M. Immunoglobulin light chain amyloidosis: 2020 update on diagnosis, 33. Chattopadhyay, A. & Haldar, S. Dynamic insight into protein structure prognosis, and treatment. Am. J. Hematol. 95, 848–860 (2020). utilizing red edge excitation shift. Acc. Chem. Res. 47,12–19 (2014). 4. Baden, E. M., Randles, E. G., Aboagye, A. K., Thompson, J. R. & Ramirez- 34. Catici, D. A. M., Amos, H. E., Yang, Y., van den Elsen, J. M. H. & Pudney, C. Alvarado, M. Structural insights into the role of mutations in amyloidogenesis. R. The red edge excitation shift phenomenon can be used to unmask protein J. Biol. Chem. 283, 30950–30956 (2008). structural ensembles: implications for NEMO-ubiquitin interactions. FEBS J. 5. Hurle, M. R., Helms, L. R., Li, L., Chan, W. & Wetzel, R. A role for 283, 2272–2284 (2016). destabilizing amino acid replacements in light-chain amyloidosis. Proc. Natl 35. Serrano, A. L., Lomont, J. P., Tu, L.-H., Raleigh, D. P. & Zanni, M. T. A free Acad. Sci. USA 91, 5446–5450 (1994). energy barrier caused by the refolding of an oligomeric intermediate controls 6. Kazman, P. et al. Fatal amyloid formation in a patient’s antibody light chain is the lag time of amyloid formation by hIAPP. J. Am. Chem. Soc. 129, caused by a single point mutation. Elife 9,1–23 (2020). 16748–16759 (2017). 7. Ramirez-Alvarado, M. Amyloid formation in light chain amyloidosis. Curr. 36. Martin, D. J. & Ramirez-Alvarado, M. Comparison of amyloid fibril formation Top. Med. Chem. 12, 2523–2533 (2012). by two closely related immunoglobulin light chain variable domains. Amyloid 8. Buxbaum, J. Aberrant immunoglobulin synthesis in light chain amyloidosis. J. 17, 129–136 (2010). Clin. Invest. 78, 798–806 (1986). 37. Blancas-Mejía, L. M. et al. Thermodynamic and fibril formation studies of full 9. Gallo, G., Wisniewski, T. & Ghiso, J. Short communication fibrillogenesis. Am. length immunoglobulin light chain AL-09 and its germline protein using scan J. Pathol. 145, 526–530 (1994). rate dependent thermal unfolding. Biophys. Chem. 207,13–20 (2015). NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 38. Weber, B. et al. The antibody light-chain linker regulates domain orientation 61. Phillips, S. R., Wilson, L. J. & Borkman, R. F. Acrylamide and iodide and amyloidogenicity. J. Mol. Biol. 430, 4925–4940 (2018). fluorescence quenching as a structural probe of tryptophan microenvironment 39. Karamanos, T. K. et al. Structural mapping of oligomeric intermediates in an in bovine Lens crystallins. Curr. Eye Res. 5, 611–620 (1986). amyloid assembly pathway. Elife 8,1–32 (2019). 62. Tallmadge, D. H., Huebnert, J. S. & Borkman, R. F. Acrylamide quenching of 40. Souillac, P. O. et al. Elucidation of the molecular mechanism during the early tryptophan photochemistry and photophysics. Photochem. Photobiol. 49, events in immunoglobulin light chain amyloid fibrillation: evidence for an off- 381–386 (1989). pathway oligomer at acidic pH. J. Biol. Chem. 277, 12666–12679 (2002). 41. Zhijie, Q., Dongmei, H., Min, Z. & Fink, A.L.Structuralcharacterizationofthe Acknowledgements partially folded intermediates of an immunoglobulin light chain leading to amyloid This work was supported by a grant from the DFG to J.B. (BU836/14-1) in the context of fibrillation and amorphous aggregation. Biochemistry 46, 3521–3531 (2007). the research unit FOR 2969. We thank Dragana Catici for help with REES analyses and 42. Oberti, L. et al. Concurrent structural and biophysical traits link with Laura Meier for her help with protein expression and purification. immunoglobulin light chains amyloid propensity. Sci. Rep. 7,1–11 (2017). 43. Crespo, R., Rocha, F. A., Damas, A. M. & Martins, P. M. A generic crystallization-like model that describes the kinetics of amyloid fibril Author contributions formation. J. Biol. Chem. 287, 30585–30594 (2012). P.K. and J.B. designed the study, P.K., H.E. and R.A. performed the experiments, P.K., 44. Chatani, E. & Yamamoto, N. Recent progress on understanding the H.E. and R.A. analyzed the results, P.K., R.A. and J.B. wrote the manuscript. mechanisms of amyloid nucleation. Biophys. Rev. 10, 527–534 (2018). 45. Feige, M. J., Hendershot, L. M. & Buchner, J. How antibodies fold. Trends Funding Biochem. Sci. 35, 189–198 (2010). Open Access funding enabled and organized by Projekt DEAL. 46. Huber, R., Deisenhofer, J., Colman, P. M., Matsushima, M. & Palm, W. Crystallographic structure studies of an IgG molecule and an Fc fragment. Nature 264, 415–420 (1976). Competing interests 47. Eisenhaber, F. & Argos, P. Hydrophobic regions on protein surfaces: The authors declare no competing interests. definition based on hydration shell structure and a quick method for their computation. Protein Eng. 9, 1121–1133 (1996). Additional information 48. Moelbert, S. & Emberly, E. Correlation between sequence hydrophobicity and Supplementary information The online version contains supplementary material surface-exposure pattern of database proteins. Protein Sci. 13, 752–762 (2004). available at https://doi.org/10.1038/s41467-021-26845-0. 49. Young, L., Jernigan, R. L. & Covell, D. G. A role for surface hydrophobicity. Protein Sci. 3, 717–729 (1994). Correspondence and requests for materials should be addressed to Johannes Buchner. 50. Hora, M. et al. MAK33 antibody light chain amyloid fibrils are similar to oligomeric precursors. PLoS ONE 12,1–14 (2017). Peer review information Nature Communications thanks Gareth J Morgan, Marina 51. Dear, A. J. et al. Kinetic diversity of amyloid oligomers. Proc. Natl Acad. Sci. Ramirez-Alvarado and the other, anonymous, reviewer(s) for their contribution to the USA 117, 12087 (2020). peer review of this work. 52. Arden, B. G. et al. Measuring the energy barrier of the structural change that initiates amyloid formation. Anal. Chem. 92, 4731 (2020). Reprints and permission information is available at http://www.nature.com/reprints 53. Sakurai, K. & Tomiyama, R. Enhanced accessibility and hydrophobicity of amyloidogenic intermediates of the β2-microglobulin D76N mutant revealed Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in by high-pressure experiments. J. Biol. Chem. 296, 100333 (2021). published maps and institutional affiliations. 54. Corazza, A. et al. Properties of some variants of human β2-microglobulin and amyloidogenesis. J. Biol. Chem. 279, 9176–9189 (2004). 55. Taricska, N. et al. The route from the folded to the amyloid state: exploring the potential energy surface of a drug‐like miniprotein. Chemistry 26, 1968 (2020). Open Access This article is licensed under a Creative Commons 56. Klein, W. L., Stine, W. B. & Teplow, D. B. Small assemblies of unmodified Attribution 4.0 International License, which permits use, sharing, amyloid-protein are the proximate neurotoxin in Alzheimer’s disease. adaptation, distribution and reproduction in any medium or format, as long as you give Neurobiol. Aging 25, 569–580 (2004). appropriate credit to the original author(s) and the source, provide a link to the Creative 57. Kihara, M. et al. Seeding-dependent maturation of β2-microglobulin amyloid Commons license, and indicate if changes were made. The images or other third party fibrils at neutral pH. J. Biol. Chem. 280, 12012–12018 (2005). material in this article are included in the article’s Creative Commons license, unless 58. Nokwe, C. N. et al. The antibody light-chain linker is important for domain indicated otherwise in a credit line to the material. If material is not included in the stability and amyloid formation. J. Mol. Biol. 427, 3572–3586 (2015). article’s Creative Commons license and your intended use is not permitted by statutory 59. Brown, P. H. & Schuck, P. Macromolecular size-and-shape distributions by regulation or exceeds the permitted use, you will need to obtain permission directly from sedimentation velocity analytical ultracentrifugation. Biophys. J. 90, the copyright holder. To view a copy of this license, visit http://creativecommons.org/ 4651–4661 (2006). licenses/by/4.0/. 60. Schuck, P. Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and Lamm equation modeling. Biophys. J. 78, 1606–1619 (2000). © The Author(s) 2021 10 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Dissection of the amyloid formation pathway in AL amyloidosis

Loading next page...
 
/lp/springer-journals/dissection-of-the-amyloid-formation-pathway-in-al-amyloidosis-ykHerLoP87

References (67)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2041-1723
DOI
10.1038/s41467-021-26845-0
Publisher site
See Article on Publisher Site

Abstract

ARTICLE https://doi.org/10.1038/s41467-021-26845-0 OPEN Dissection of the amyloid formation pathway in AL amyloidosis 1,3 1,3 2 1 Pamina Kazman , Ramona M. Absmeier , Harald Engelhardt & Johannes Buchner In antibody light chain (AL) amyloidosis, overproduced light chain (LC) fragments accumu- late as fibrils in organs and tissues of patients. In vitro, AL fibril formation is a slow process, characterized by a pronounced lag phase. The events occurring during this lag phase are largely unknown. We have dissected the lag phase of a patient-derived LC truncation and identified structural transitions that precede fibril formation. The process starts with partial unfolding of the V domain and the formation of small amounts of dimers. This is a pre- requisite for the formation of an ensemble of oligomers, which are the precursors of fibrils. During oligomerization, the hydrophobic core of the LC domain rearranges which leads to changes in solvent accessibility and rigidity. Structural transitions from an anti-parallel to a parallel β-sheet secondary structure occur in the oligomers prior to amyloid formation. Together, our results reveal a rate-limiting multi-step mechanism of structural transitions prior to fibril formation in AL amyloidosis, which offers, in the long run, opportunities for therapeutic intervention. 1 2 Department Chemie, Technische Universität München, 85748 Garching, Germany. Department Molecular Structural Biology, Max-Planck-Institute of 3 ✉ Biochemistry, 82152 Martinsried, Germany. These authors contributed equally: Pamina Kazman, Ramona M. Absmeier. email: johannes.buchner@tum.de NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 16,23 ight chain (AL) amyloidosis is the most common type of cooperative reaction as monitored by ThT fluorescence .The systemic amyloidosis . The disease is caused by an under- molecular events occurring in this lag phase are still elusive. For a Llying plasma cell dyscrasia that entails the elevated expres- better understanding of AL amyloidosis, we set out to resolve the 2,3 sion and secretion of free antibody light chains (LC) .Patient- molecular mechanism of structural changes occurring during the lag specificmutations in theLC, whichdestabilize thenativefoldand phase for the well-characterized patient LC truncation Pat-1 that consequently trigger fibril formation are an important element of the represents the major component of the deposited fibrils. The struc- 4–7 disease . Furthermore, the circulating free LCs often undergo ture and amyloidogenic properties of Pat-1 had been reported proteolytic cleavage prior to assembly into insoluble amyloid previously . In this study, two disease-causing mutations (L15P 8–11 fibrils .N-terminalfragments,comprisingmostlythe variable L82Q) had been identified in Pat-1 compared to the germline domain (V ) of the LC often represent the amyloidogenic species. sequence, which did not form fibrils. For the lag phase analysis, we The hallmark of AL amyloidosis is the transformation of soluble used the Pat-1 V , the double mutant Pat-1 L15P L82Q and the 12,13 monomeric protein into insoluble amyloid fibrils . Their presence respective germline sequence WT-1. To obtain insights into the correlates strongly with the diseaseand theimpairmentoforgansin molecular events occurring during the lag phase, we first monitored which they are deposited. Structurally, substantial rearrangements of changes in the solubility of the V domain under conditions favoring the V domain in AL fibrils compared to the native fold have rapid fibril formation (37 °C, shaking at 750 rpm, 0.5 mM SDS). Low occurred as resolved by cryo-EM. During restructuring, the internal concentrations of SDS are typically used as a destabilizing agent to 6,24 disulfide bond is retained and hydrophobic core residues become accelerate fibril formation of amyloid proteins .Aliquotswere 14,15 surface-exposed . Like in other amyloid diseases, the growth taken at different timepoints during incubation and separated into phase of AL fibrilsisprecededbyacomparably long and rate- soluble and insoluble fractions by centrifugation. Analysis of the 16–18 limiting lag phase in vitro . So far, little is known about the fractions by SDS-PAGE and quantification of the bands showed that structural events taking place during this phase. For different amyloid during the first hour of incubation, the amount of soluble V domain diseases, oligomeric intermediates of the amyloidogenic proteins and did not change significantly (Fig. 1a). Then, a decrease of soluble Pat- 16,19–22 a nucleated polymerization mechanism have been suggested . 1 and a concomitant increase of Pat-1 in the insoluble fraction were Capturing these species remains challenging, due to their transient observed over time. The half times of the reactions were very similar, appearance in low concentrations and the high energy states of the with a t of 1.79 h ± 0.11 for the decrease of the soluble form and t 1/2 1/2 16,18 specificintermediates . In AL amyloidosis, the presence of oli- of 1.71 h ± 0.30 h for the increase in the insoluble species (Fig. 1b). gomers forming during the lag phase and the accompanied structural After 2 h, almost no protein was left in the soluble fraction. Fibril transitions prior to amyloid formation have not been investigated so formation was confirmed by the measurement of the ThT fluores- far. Understanding the pathway and molecular mechanism of reac- cence. Fibrils started to form after 2 h and the half time of the tions preceding fibril formation of pathogenic LCs is important to reaction was 3.5 h (Fig. 1b).After 4h,nofurther increase in ThT identify potential therapeutic intervention points at early stages of the fluorescence was visible, thus no further fibril growth seems to take disease. Here, we elucidated the processes taking place during the lag place while ongoing structural rearrangement cannot be excluded. phase prior to fibril formation of the well-studied pathogenic V The presence of fibrils was confirmed by TEM micrographs (Fig. 1c). domain Pat-1 .Weidentified intermediate oligomeric species on the For Pat-1 L15P L82Q, fibril formation occurred even faster while the fibril pathway and associated structural rearrangements using a broad germline did not form fibrilsasconfirmed by ThT fluorescence and range of biophysical analyses. TEM (Supplementary Figs. 1, 4d, h). When we transferred the Pat-1 at different timepoints during the lag phase to lower temperatures (4 °C or 20 °C) and stopped shaking, ThT fluorescence analysis Results showed that no further fibril formation occured over a time period of Fibril formation is preceded by the disappearance of soluble V 3 h (Supplementary Fig. 2). This allowed us to dissect the lag phase monomers.In AL amyloidosis, fibril formation in vitro is usually and perform detailed analyses of the samples. preceded by a long lag phase after which fibrils form in a rapid and Fig. 1 Transition from soluble V Pat-1 to insoluble fibrils. a Soluble and insoluble fractions of Pat-1 V during incubation at amyloid-promoting conditions. L L Samples were taken at the timepoints indicated and run on an 18% SDS-gel. b Quantified and normalized band intensities of the SDS-gel are shown in turquoise for the soluble V domain (t = 1.79 h) and in black for the insoluble fraction (t = 1.71 h). Shades represent the SEM of n = 2. Fibril formation L 1/2 1/2 of Pat-1 V as monitored by ThT fluorescence is shown in pink (t = 3.48 h). Values were fit to a Boltzmann function Shades represent the SEM of n = 3. L 1/2 c TEM micrograph of fibrils formed after 3 h of incubation. The scale bar represents 200 nm. 2 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE Fig. 2 Oligomeric species present in the lag phase of the amyloid pathway of Pat-1 V . AUC sedimentation profiles over time of a monomeric and dimeric species and b oligomeric species. c Normalized AUC peak quantification of monomeric (yellow) and oligomeric (purple) species during the lag phase of fibril formation. The data for monomeric species were fit to a Boltzmann function, the data for oligomeric species were fit to a Gaussian function. Shade of the ThT Fluorescence represents the SEM of n = 3. d Section of TEM micrographs shown in Fig. S5 of the oligomer and fibril formation over time. Scale bar represents 50 nm. Different oligomeric intermediates are part of the pathway to hour of incubation. After that, they cannot be detected unam- fibril formation. Due to the denaturing conditions, SDS-PAGE biguously due to the lower concentration of monomers and the analysis does not provide information on noncovalent structural lack of resolution provided by the fitting software. TEM micro- alterations, specifically oligomerization, taking place during the graphs confirmed the presence of oligomeric species after 15 min lag phase. Thus, to determine potential changes in quaternary incubation of Pat-1 and immediately before fibril formation after structure, we performed analytical ultracentrifugation (AUC). For 1.75 h in Pat-1 L15P L82Q. Over time, the oligomer amount a better resolution of the different species we increased the pro- increases and fibrils for Pat-1 could be detected after 3 h and after tein concentration. The sedimentation analysis revealed that at 2 h for Pat-1 L15P L82Q consistent with the results of the ThT the start point (0 h), Pat-1 was present mainly as a monomer with assay (Fig. 2d, Supplementary Figs. 1, 4d, 5). Large oligomers a small fraction of dimers (Fig. 2a, Supplementary Fig. 3). After were visible in the TEM micrographs in a lower number, which 15 min incubation, a range of oligomers became visible as four could be due to their dissociation under the acidic conditions distinct peaks in the range from 4 to 20 S in addition (Fig. 2b, during negative staining. The germline formed oligomeric species Supplementary Fig. 3). The amount of oligomeric species with a clustering morphology. However, here no fibrils but increased during the first 30 min, while the monomer fraction amorphous aggregates were formed (Supplementary Fig. 4e-h). decreased. After 45 min, the oligomeric species reached a peak, while about half of the sample was still monomeric. After that The V secondary structure changes during oligomerization. timepoint, both oligomers and monomers decreased and dis- As the native V domain and AL fibrils differ substantially in side appeared completely, and fibrils evolved (Fig. 2a–c). At each 14,15 chain interactions , conformational remodeling of V has to timepoint tested, the concentration of oligomers was lower than occur in the lag phase. To detect potential secondary structure the concentration of monomers, as determined by the area under changes, we followed the reaction over time by far UV (FUV) the curves (Fig. 2a–c). For Pat-1 L15P L82Q, the appearance of circular dichroism (CD) spectroscopy. At the start of the reaction, oligomers before fibril formation was also observed (Supple- Pat-1 V showed a minimum at 218 nm followed by an amplitude mentary Fig. 4a, b). Since the monomer peak (1.5 S) shows a L as expected for the antiparallel β-sheet native structure of Ig shoulder towards higher S values (2.1 S) from the beginning of the domains (Fig. 3a) . After 0.5 h incubation, the amplitude of the incubation onwards, we conclude that initially V dimers form in minimum at 218 nm increased. The alterations observed during Pat-1 and Pat-1 L15P L82Q (Fig. 2a, Supplementary Figs. 2, 4a-c). the lag phase indicate a reorientation and partial unfolding of These dimers remain present in low amounts during the first NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 Fig. 3 Secondary structure changes during oligomer formation monitored by FUV CD-and FTIR spectroscopy. a Comparison of changes in the secondary structure of Pat-1 in the lag phase. The spectra show the average of 10 individual scans. b Secondary structure analysis of the CD spectra of Pat-1 using the BeStSel algorithm. c Secondary structure changes in the lag phase monitored by ATR-FTIR after H/D exchange. The spectra of the amide I and II region were scaled to an absorbance of 1 and baseline corrected. d Comparison of the shift of the minimum in the spectra of different timepoints during the lag phase (blue), the change of intensity at 210 nm (green) with the fibril formation kinetics followed by ThT fluorescence (pink). Shades of the ThT fluorescence represent the SEM of n = 3. secondary structure elements. This trend continued further dur- domain rearrangements during oligomer formation and rear- ing 1.5 h of incubation. After 1.75 h, the CD minimum shifted to rangement. Upon fibril formation, the signal remains constant. 220 nm and the amplitude increased. At the same time, a max- Taken together, the CD analysis suggests that despite the imum at 204 nm emerged. The shift of the minimum and the apparent rearrangement of secondary structure, secondary simultaneous intersection of the x-axis at higher wavelengths structure elements remain present during the lag phase. Con- indicates the formation of parallel β-sheets . Thus, this time- comitant with fibril formation, the CD data suggest a structural point marks an important conformational rearrangement. During rearrangement event leading from antiparallel to cross β-sheets. the lag phase, the high voltage (HT) at the detector, which reflects To further analyze secondary structure changes during the lag the absorption of light by the sample, also changed. The signal phase we used attenuated total reflection Fourier transform decreased during the lag phase, which means more light reaches infrared spectroscopy (ATR-FTIR). In order to minimize the detector (Fig. S6). This in turn suggests that structural contributions in the Amid I region originating from residual changes affecting the signal take place. After 2 h, the minimum water bound to buffer components we recorded spectra after H/D shifted further to 224 nm and the maximum at 204 nm was more exchange (Supplementary Fig. 7c). The overall content of β- pronounced. This maximum is indicative of supramolecular β- structure remains unchanged in samples of mono- and oligomers sheet-rich amyloid structures . A similar effect was also observed while fibril formation results in a peak shift from 1638 to 1625 cm −1 for Pat-1 L15P L82Q but not for the germline WT-1 (Supple- because of intermolecular β-sheet stacking in the fibrillar mentary Fig. 7a). Analysis of the CD data with the BeStSel structure (Fig. 3c) . Concomitantly, the shoulder around −1 algorithm revealed an decrease of antiparallel β-strands with a 1685 cm disappears. This observation indicates a decrease of 30 −1 slight increase in parallel β-strands and the unfolded fraction for antiparallel β-sheet content . The peaks at 1590 and 1515 cm Pat-1 and Pat-1 L15P L82Q but not for WT-1 (Fig. 3b, Supple- mainly belong to side chain absorptions and Tyr, respectively . mentary Fig. 7b). The shift of the minimum in the FUV CD spectra over time precedes the appearance of fibrils, as the ThT fluorescence starts to increase after the wavelength shift of the CD The V domain structure changes during oligomerization.To minimum to 224 nm (Fig. 3d). Interestingly, even before the shift monitor structural changes in the core of the V domain, we in the CD minimum occurs, the signal of the amplitude at lower analyzed the fluorescence of the single intrinsic tryptophan as a wavelengths decreases in the first hour, which might indicate sensitive and specific probe since its emission intensity is quen- partial unfolding. The changes are particularly noticeable at a ched in the native state by the disulfide bridge located in the core wavelength of 210 nm, which is why these were used for com- of the β-barrel and becomes higher upon unfolding . Thus, parison in Fig. 3d. The following increase of the amplitude conformational changes involving the protein core can be mon- between 1 and 2 h of the lag phase seems to correspond to itored by a change in the fluorescence amplitude of tryptophan. 4 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE Fig. 4 Structural changes of Pat-1 during the lag phase. a Change of tryptophan fluorescence. Each emission spectrum is the average of three individual scans. b Change of surface hydrophobicity followed by ANS fluorescence. c Change of tryptophan fluorescence upon quenching with 0–1 M acrylamide. The curves show the respective F /F values of the Stern-Volmer equation, Error bars represent the SEM of n = 3. d Comparison of the normalized change of ANS fluorescence (gray), Trp exposure (orange) and the quenching constant K of the acrylamide quenching calculated by Eq. (1) (purple) in comparison to changes in the ThT signal (pink). Shades represent the SEM of n = 3. When we analyzed changes in the fluorescence emission during detected by CD spectroscopy (Figs. 3a, 4b; Supplementary the lag phase, we observed a steady increase from the beginning Fig. 8c). onwards indicating that structural rearrangements take place in To extend our analyses of conformational rearrangements via the protein core (Fig. 4a). The fluorescence intensity reached its the accessibility of tryptophan, fluorescence quenching by maximum after 3 h, at the same time when fibrils have formed acrylamide was assessed. The quenching is stronger the more (Fig. 4d). Pat-1 L15P L82Q and WT-1 also showed an increase in solvent accessible the tryptophan residue is. There is a steep tryptophan fluorescence which occurred rapidly upon fibril for- increase in fluorescence quenching already at the beginning of the mation for the double mutant while the germline shows no fibril lag phase and the amplitude continues to grow up to 1 h. This formation (Supplementary Figs. 4, 8a, b). indicates that the structural rearrangement we observed via To further probe structural changes of the domain surface, we changes in tryptophan fluorescence involves the rapid reposition- used the fluorophore ANS which binds specifically to surface- ing of the buried tryptophan to a solvent-exposed position exposed hydrophobic patches in a protein. Upon binding, the (Fig. 4c). fluorescence intensity increases and the emission maximum shifts To obtain further structural insight in the changes of the to lower wavelengths . Following ANS fluorescence during the tryptophan environment, we employed red edge excitation shift lag phase, we observed an increase in ANS fluorescence intensity (REES) spectroscopy . The REES effect is driven by the dipole from the beginning of the incubation until a maximum was interactions of the fluorophore with its surrounding: a rigid or a reached after 2 h. The increase in surface hydrophobicity occurs completely solvent-exposed surrounding leads to a smaller effect for both, Pat-1 L15P L82Q and the germline, although to a much than a flexible protein present in different conformational greater extent in the double mutant (Supplementary Figs. 8c, d). states . For the Pat-1 V domain, we observed an increase of Since the germline also forms oligomers and aggregates, an the center of spectral mass (CSM) after 0.25 h compared to the increase in ANS binding is reasonable. Thus, our results indicate start of the reaction (Fig. 5a, b). This implies a strong increase in that rearrangements in the environment of the tryptophan in the solvent exposure of the tryptophan at the beginning of the lag core also leads to changes in the surface hydrophobicity in Pat-1 phase due to a domain opening. After 0.25 h, a second, slower and its double mutant. This domain opening and increasing phase became apparent, which increased over time (Fig. 5B). The solvent exposure of the inner core of the VL domains occurs first higher REES effect suggests a rugged free energy landscape of the in the lag phase, followed by changes in secondary structure as V domain at the beginning of the reaction, which decreases NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 Fig. 5 The REES effect of Pat-1 V during the lag phase. a Change in the CSM for the corresponding excitation wavelengths from 280 to 300 nm. b Conclusion of the REES effect (orange) and solvent exposure (black) at the different timepoints. Measurements were carried out in triplicates. Error bars respresent the respective SEM of the triplicates. quickly, indicating that there are less conformational states phase, a specific pattern of higher oligomers was detected which available, which influence tryptophan fluorescence relaxation. rearranged before the assembly into fibrils starts. Concomitantly, The smaller REES effect is not due to an extensive unfolding of the monomers decrease during the lag phase. The oligomer the protein, as in the presence of 6 M urea a pronounced shift to fraction shows a complex behavior that slightly differs between smaller values for the REES effect was observed (Supplementary AUC and TEM analysis. According to AUC analysis, the amount Fig. 9c). Pat-1 L15P L82Q shows a similar effect concerning of oligomers first increases and then decreases until they com- increased solvent exposure and decreasing REES effect (Supple- pletely disappear when fibrils are formed. This hints at the for- mentary Fig. 9c). The later stays the same for the germline while mation of a nucleus prior to fibril formation as proposed for other 43,44 also an increase in the solvent exposure can be observed matching amyloid reactions . The cooperative transition to the fibrillary the fluorescence data (Supplementary Figs. 8b, 9b). Thus, the two state supports the view that a specific conformational state of the phases of the REES kinetics during the lag phase may be related to oligomers is the seed for polymerization. Accordingly, the oli- the initial partial unfolding with a concomitant higher solvent gomeric species were always present in lower amounts compared accessibility or a higher rigidity due to dimerization followed by to soluble monomer/dimer fraction or insoluble fibrils as oligomer formation and rearrangement in the oligomer. In described for other amyloids . TEM micrographs of Pat-1 contrast to the fibril forming proteins, the germline accesses a showed oligomers that could correspond to hexamers, which lower amount of conformational states as indicated by the overall increase during the lag phase in number but not in size; the Pat-1 lower REES effect (Supplementary Fig. 9c) double mutant L15P L82Q formed similar oligomers rapidly before fibril formation. As negative staining involves the incu- bation in a low pH uranyl acetate solution, we assume that this Discussion dissolves the higher oligomers. Thus, the hexamer is the most The conformational switch from the native to the fibrillary state stable oligomeric species. In contrast to Pat-1, the WT-1 oligomer in amyloidosis comprises a lag phase, including a primary species exhibited clustering and unspecific aggregation in TEM nucleation step, a transition phase in which fibrils start to form analysis. and elongate, and a final plateau phase in which fibrils are present 16,35 According to the known secondary and tertiary structure of the and an equilibrium is reached . The same general scheme was 6,36–38 monomers as well as the fibrils, major conformational rearran- assumed to apply for AL amyloidosis . However, the gements have to occur in V prior to fibril formation as the molecular events occurring in early phases of the fibrillation domain consists of a two-layer sandwich structure composed of process were largely unknown. In this study, we determined the antiparallel β-strands, whereas the amyloid fibrils in AL amyloi- conformational transitions and the molecular species formed in dosis exhibit a cross-β sheet topology consisting of parallel β- the lag phase that predispose the pathogenic V domain to fibril 14,15,45,46 sheets . The single tryptophan residue buried in the core formation. of the domain is an excellent spectroscopic probe for con- We show that, starting from the monomeric patient LC trun- 6 formational changes as it reports on variations in its local cation Pat-1 , small amounts of dimers are formed in the first environment . Analysis of the intrinsic tryptophan fluorescence hour of the lag phase. They might be caused by a dynamic revealed that the microenvironment of the tryptophan residue in equilibrium between these two states. Later on, the peak shoulder the core starts to change already in the very beginning of the lag is not readily visible, however, we presume that a fraction of phase as seen by an increase of overall fluorescence intensity. This dimers is still present. As the dimeric fraction is the first obser- effect increases with time indicating that the domain structure vable step that differs between native and fibril-inducing condi- increasingly changes. The distancing of the tryptophan residue tions, these dimers reflect conformational changes which lead to 6 from the disulfide bond and the enhanced accessibility reflects the destabilized monomers which finally form fibrils . Of note, we opening of the β-barrel structure while the overall secondary consider these dimers as non-native species induced by the structure elements still remain largely intact as judged from the increasing hydrophobic surface at a beginning unfolding. They FUV CD measurements. This domain opening goes along with need to be differentiated from the native dimers observed for a the formation of small amounts of dimers as seen in the AUC number of LCs. In this context, the dimers either do not interfere 36,39 sedimentation profiles. Thus, we hypothesize that even though with fibril formation or they may exert a protective role 40–42 the overall secondary structure of the V monomer does not against fibrillization . In addition, after 0.25 h of the lag 6 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE change during initial oligomerization, a state appears in which the Compared to the disease-related V domains, the germline packing starts to rearrange, and hydrophobic residues become behaves differently in important aspects. While it also undergoes surface-exposed. ANS binding experiments show an early partial unfolding that leads to an increase in the tryptophan increase of surface-exposed hydrophobicity, which seems to fluorescence and a slight increase in surface hydrophobicity, the coincide with domain opening and oligomer formation. At this conformational state of the germline remains constant during the step, a further conformational reorganization occurs that fosters lag phase and there are no observable changes in the secondary oligomerization via hydrophobic interactions, as observed for structure. The oligomeric species formed end in amorphous 47–49 other amyloidogenic proteins . The acrylamide quenching aggregates. experiments give additional support to the idea of conformational In summary, our biophysical analyses reveal a multi-step rearrangements during oligomer formation. It monitors the conformational transition from a folded monomeric β-barrel accessibility of a quencher to a fluorophore and thus is a marker domain into an amyloid fibril (Fig. 6). It starts with an initial for solvent accessibility. The increase of the quenching constant K domain opening and partial unfolding in which the β-strands of from the beginning of the lag phase onwards implies a higher the V monomers are preserved. The increase in surface hydro- solvent exposure of the buried tryptophan along with the partial phobicity fosters dimerization and concomitant assembly into unfolding of the V domain. REES experiments further revealed hexamers and multiples of hexamers. In this process further that an increased solvent accessibility of the tryptophan residue, rearrangements occur, which reduce conformational flexibility. monitored by the CSM at an excitation wavelength of 280 nm, Consistent with the idea that the oligomers are the species with takes place. This again supports the notion of domain opening the highest Gibbs free energy, they are present only transiently concluded from changes in tryptophan fluorescence and ANS and at low concentration. It was previously reported that the life binding. The increase of CSM values at the y-intercept goes along time of oligomers of amyloid proteins can vary significantly. Also, with a decrease in the REES effect. The emission spectrum of a oligomers may have a higher tendency for dissociation to fluorophore is highly dependent on its environment since it is monomers than for fibril formation . influenced by the dipole interaction with water. A more solvent- In the context of our study, it is useful to compare the concept exposed surrounding thereby leads to a smaller REES effect, as established for AL amyloidosis with other fibril forming proteins. shown by the unfolding of Pat-1 with urea. The magnitude of the The dialysis-related amyloid precursor β -microglobulin also REES effect can provide information about the free energy consists of an immunoglobulin fold. It was shown that starting landscape of a protein . A higher REES effect is based on a from the natively folded monomer, an unfolded state emerges 39,52–54 higher number of discrete conformational states as seen for Pat-1 and subsequent oligomerization precedes fibril formation . and Pat-1 L15P L82Q. Upon oligomerization this REES effect Clustering following unfolding has also been suggested for amy- decreases, which also implies a decrease in conformational states. loidosis involving α-helical precursors . Our study shows that In contrast, WT-1 shows unfolding events but the accessible the decisive committing step for the amyloid pathway occurs in conformational states do not increase. Since the fluorescence the context of the oligomer. In contrast, unfolding and initial intensity does not decrease during the lag phase and during fibril oligomerization were also observed for the nonamyloidogenic formation, the tryptophan does not shift back to its position in germline protein. However, this protein subsequently ends up in close proximity of the quenching disulfide bond. These results are amorphous aggregates. Thus, it seems that only once the oligo- in excellent agreement with the cryo-EM structure of AL fibrils, mers undergo a critical structural rearrangement nuclei are where the conserved tryptophan residue is found close to the formed, which are rapidly transformed into amyloid fibrils. For fibril surface. Furthermore, the disulfide bond is intact but the further insights into the structure of the oligomers, high resolu- interactions of residues including the tryptophan differ sig- tion methods like cryo-EM or solid state NMR need to be applied 14,50 nificantly from that in the native protein . We assume that in the future. Especially, since pathogenic effects could be ascri- after initial unfolding events and partial dimer formation from bed not only to insoluble fibrils but also to oligomers . Fur- folded monomers, a subsequent fast association of monomeric or thermore, a molecular understanding of the conformational dimeric species to oligomers occurs. Preceding fibril formation transitions in the lag phase of different amyloidoses may result in but after the initial unfolding events, a shift in the FUV CD the emergence of general concepts across diseases. minima became visible which we assume represents the rear- rangement of the β-sheets. The change of local minima fits to the Methods switch of antiparallel to parallel β-sheet fold . Thus, the CD All chemicals were from Merck (Darmstadt, Germany) or Sigma (St. Louis, USA). results indicate that this timepoint marks further important All measurements were carried out in PBS buffer containing small amounts of structural rearrangements induced by intermolecular interactions. 24,57,58 SDS (10 mM Na HPO ×2 H O; 1.8 mM KH PO ; 2.7 mM KCl; 137 mM 2 4 2 2 4 The rapid reaction into fibrillary structures with features of a NaCl; 0.5 mM SDS) at pH 7.4 and 37 °C, unless otherwise stated. Data were supramolecular β-sheet formation reveals that the rearranged analyzed using Origin 2019. species is transient, meta-stable and thus potentially presents the nucleus for fibril formation. The β-sheet formation in fibrils was Expression and purification of Pat-1, Pat-1 L15P L82Q, and WT-1. The origin of also observed in FTIR, the rapid switch after ~2 h could not be the Pat-1 sequence and the recombinant expression and purification of Pat-1, Pat-1 resolved in detail. However a difference in the amide II region L15P L82Q, and WT-1 was described before. Also, the generation of the point emerged. Of note, the FTIR measurements required extensive mutations with the pirmers TAGCGGTAGCCCGGGTCAGAGCATTA (+) and ACGCTTGCAGGCTGGGTC (−) has been previously described. In brief, the sample processing and could not be conducted in a time-resolved plasmids were transformed in E. coli BL21 (DE3)-star cells and protein expression manner similar to the CD. In this context, it should be noticed took place at 37 °C overnight. Cells were harvested and inclusion bodies were that in general conditions were kept identical between the dif- prepared. The pellet was solubilized and unfolded in 25 mM Tris-HCl (pH 8), ferent methods used. However, additives like ANS or ThT, or 5 mM EDTA, 8 M urea, and 2 mM β-mercaptoethanol at room temperature for a slightly different conditions like in the AUC experiments might minimum of 2 h. Afterwards, the protein was loaded onto a Q-Sepharose anion exchange column equilibrated in 25 mM Tris-HCl (pH 8.0), 5 mM EDTA and 5 M potentially impact the conformational equilibrium. However, the urea. The LCs and V s were eluted in the flow-through fractions and refolded by methodologies have been well established for the addressed dialysis against 250 mM Tris-HCl (pH 8.0), 100 mM L-Arg, 5 mM EDTA, 1 mM questions and the results obtained are in line with each other, oxidized glutathione and 0.5 mM reduced glutathione at 4 °C overnight. To remove supporting the individual evidence. aggregates and impurities, the refolded proteins were purified using a Superdex 75 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 Monomers/Dimers Oligomers Fibrils rearrangement Time Folded Monomer Oligomer Oligomer from Fibrils monomer with partially with partially structually rearranged opened structure opened monomers monomers - nucleus Fig. 6 Model of the events taking place in the lag phase of fibril formation. Folded V monomers undergo a partial structural opening whereupon they assemble to dimers, and oligomers (hexamers and multiples of hexamers). Prior to fibril formation, the monomers in these oligomers structurally rearrange resulting in parallel β-sheets. This variant is the nucleus of fibrillization. The above bars show the concentration of each species at the respective timepoint of the fibril formation process. The concentration of oligomers is low at all timepoints. 16/60 gel-filtration column (GE Healthcare, Uppsala, Sweden) equilibrated in PBS 70 from Bruker, Germany). The samples were equilibrated in the measurement buffer. Recovery and purity of intact proteins were analyzed by SDS-PAGE. chamber under nitrogen flow for 30 min and subsequently incubated with D O saturated nitrogen flow for H/D exchange of bound water fractions. The kinetics of −1 H/D exchange was repeatedly recorded with 32 scans at 2 cm resolution until the Sample preparation for lag phase analysis. For studying the different time- spectra remained unchanged. After 50 min the final spectra were recorded, accu- points, for all measurements unless otherwise stated, 15 µM of Pat-1 were prepared mulating 1024 scans. The linear baseline was subtracted from spectra for further in the prior described assay buffer in a 1.5 mL Eppendorf reaction tube and were analyses and documentation. incubated at 37 °C ± 2 and 750 rpm shaking in a Thermomixer compact (Eppen- dorf, Hamburg, Germany). For 0 h timepoints, the sample was put on 37 °C and removed before shaking. All other timepoints were removed after the corre- 8-Anilino-1-naphtalenesulfonic acid (ANS) and Tryptophan Fluorescence. For measurement of ANS binding, samples were taken at different timepoints and sponding time shaking. All samples were kept on ice afterwards. To confirm the interruption of fibril formation at the different timepoints the shaking was stopped incubated with 150 µM ANS for 1 h. Spectra were recorded from 400 to 650 nm with an excitation of 380 nm. For intrinsic Tryptophan fluorescence measurements and the samples were kept on ice or at 25 °C for 3 h prior to ThT fluorescence analysis. spectra were recorded from 300–450 nm with an excitation wavelength of 280 nm. All fluorescence measurements were carried out using a Jasco FP-8500 Spectro- fluorometer (JASCO, Pfungstadt, Germany) at 25 °C. The settings included exci- Analytical ultracentrifugation (AUC). AUC measurements were carried out tation and emission bandwidth of 5 nm each, 4 s response time, a data interval of using an Optima AUC (Beckman, Krefeld, Germany) equipped with absorbance 1 nm and 200 nm/min scan time.. For Pat-1 L15P L82Q and WT-1 spectra were optics. The protein concentration for the measurements was 30 µM due to a low recorded at a Tecan Infinite 200 PRO M Nano with a data interval of 1 nm and an data resolution at lower concentrations. A total volume of 350 µL per sample was amplification of 100. Depicted spectra show the average of three individual mea- loaded into assembled cells with quartz windows and 12 mm-path-length charcoal- surements. Samples were taken from a 15 µM protein solution incubated at 37 °C filled epon double-sector centerpieces. The measurements were performed at and 700 rpm. 42,000 rpm in an eight-hole Beckman-Coulter AN50-ti rotor at 20 °C. Sedi- mentation was continuously scanned with a radial resolution of 10 µm and mon- Acrylamide quenching. For Acrylamide quenching, samples were taken at dif- itored at 280 nm. Data analysis was carried out with software SEDFIT using the 59,60 continuous c(S) distribution mode . ferent timepoints and 0 M to 1 M acrylamide in 1 × PBS in 0.2 M steps were added to 5 µM of the protein. The tryptophan fluorescence was recorded from 300 to 400 nm with an excitation of 280 nm. The measurement was carried out at 37 °C at Far-UV (FUV) circular dichroism (CD) measurements. For Pat-1, FUV CD a Tecan Infinite 200 PRO M Nano with an amplification of 157 and a data spectra were recorded from 197–260 nm using a Chirascan-plus CD spectrometer interval of 1 nm. The Stern-Volmer quotient F /F was calculated with the maximal (Applied Photophysics, Leatherhead, England). Measurements were recorded with fluorescence intensity at 332 nm, while F is the value at 0 M and F the fluorescence a bandwidth of 1.0 nm in 1.0 nm steps and 0.5 s time per point at a temperature of 61,62 intensity at the respective acrylamide concentration . The raw data was linearly 37 °C. All measurements were performed using a 15 µM protein solution in a fitted. The slope represents the quenching constant regarding Eq. (1): quartz cuvette with 1 mm pathlength. The spectra show an average of 10 individual measurements. For Pat-1 L15P L82Q and WT-1 FUV CD spectra were recorded from 197 to ð1Þ 260 nm using a Jasco J-1500 (JASCO, Pfungstadt, Germany). Measurements were ¼ 1 þ KQjj recorded with a data pitch of 0.1 nm and a scanning rate of 20 nm/min time per point at a temperature of 37 °C. All measurements were performed using a 15 µM protein solution in a quartz cuvette with 0.5 mm pathlength. The spectra show an average of 20 individual measurements. Red edge excitation shift (REES). The REES effect at different timepoints was conducted at a Jasco FP-8500 Spectrofluorometer at 25 °C with a scanning rate of Attenuated total reflection Fourier transform infrared spectroscopy (ATR- 500 nm/min. The setup included a bandwidth of 5 nm for excitation and emission, FTIR). After the respective timepoint of the lag phase, samples were dialyzed in a response time of 2 s and an interval of 1 nm. The tryptophan emission was 10 mM Na HPO ×2 H O; 1.8 mM KH PO overnight and 250 µl were dried on monitored from 315 to 400 nm with an excitation scan from 280 to 300 nm. For 2 4 2 2 4 one side of a germanium crystal under nitrogen flow. The crystal was mounted into data analysis, the center of spectral mass (CSM) was calculated as reported a home-made, gas-tight holder and the latter in the FTIR spectrometer (VERTEX previously . 8 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications Fibril Formation Conc. NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 ARTICLE Thioflavin T (ThT) assay. Fast fibril formation assays followed by ThT fluores- 10. Glenner, G. G., Harbaugh, J., Ohms, J. I., Harada, M. & Cuatrecasas, P. An cence were performed in 1.5 mL Eppendorf tubes. For Pat-1, kinetics were followed amyloid protein: the amino-terminal variable fragment of an immunoglobulin by measuring at different timepoints in a 10 × 2 mm quartz cuvette using a light chain. Biochem. Biophys. Res. Commun. 41, 1287–1289 (1970). FluoroMax-4 spectrofluorometer (Horiba Jobin Yvon, Bensheim, Germany) and a 11. Stevens, F. J. et al. A molecular model for self-assembly of amyloid fibrils: Jasco FP-8500 Fluorescence Spectrometer (JASCO, Pfungstadt, Germany). Kinetics immunoglobulin light chains. Biochemistry 34, 10697–10702 (1995). were followed by measuring at 440 nm excitation and 480 nm emission with slit 12. Merlini, G. & Bellotti, V. Molecular mechanisms of amyloidosis. N. Engl. J. widths of 3 nm or 2.5 nm for excitation and 4 nm or 5 nm for emission, respec- Med. 349, 583–596 (2003). tively. For Pat-1 L15P L82Q and WT-1, kinetics were followed by measuring at 13. Pepys, M. B. Amyloidosis. Annu. Rev. Med. 57, 223–241 (2006). different timepoints in a 96-well UV star plate (Greiner Bio-One, Frickenhausen, 14. Radamaker, L. et al. Cryo-EM structure of a light chain-derived amyloid fibril Germany) at a Tecan Infinite 200 PRO M Nano with an amplification of 60 and a from a patient with systemic AL amyloidosis. Nat. Commun. 10,1–8 (2019). data interval of 1 nm, The ThT fluorescence was measured from 460 to 500 nm 15. Swuec, P. et al. Cryo-EM structure of cardiac amyloid fibrils from an with an excitation wavelength of 440 nm. The data were plotted by using the ThT immunoglobulin light chain AL amyloidosis patient. Nat. Commun. 10,1–9 fluorescence at 480 nm. To remove oligomeric species and aggregated protein and (2019). prevent seed formation, monomer isolation was performed prior to the experiment 16. Arosio, P., Knowles, T. P. J. & Linse, S. On the lag phase in amyloid fibril TM by ultracentrifugation in an Optima MAX-E ultracentrifuge, (Beckman, Krefeld, formation. Phys. Chem. Chem. Phys. 17, 7607–7618 (2015). Germany). 17. Harper, J. D. & Lansbury, P. T. Models of amyloid seeding in Alzheimer’s disease and scrapie: mechanistic truths and physiological consequences of the Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). SDS time-dependent solubility of amyloid proteins. Annu. Rev. Biochem. 66, Gels were cast in two steps pouring first the 18% separating gel solution and then 385–407 (1997). the 5% stacking gel solution. For sample preparation, samples taken from 15 µM 18. Monsellier, E. & Chiti, F. Prevention of amyloid-like aggregation as a driving Pat-1 solution at different timepoints were centrifuged and separated into super- force of protein evolution. EMBO Rep. 8, 737–742 (2007). natant and pellet fraction. The pellet was washed twice with PBS. Supernatant 19. Lesné, S. et al. A specific amyloid-β protein assembly in the brain impairs samples were diluted with 5 × reducing Laemmli buffer, pellets were dissolved in memory. Nature 440, 352–357 (2006). 1 × reducing Laemmli buffer. Both were heated at 95 °C for 5 min for denaturation. 20. Kastritis, E. & Dimopoulos, M. A. Recent advances in the management of AL 20 µL of the respective samples were applied into the gel pockets. The gel was run Amyloidosis. Br. J. Haematol. 172, 170–186 (2016). at 30 mA for 40 min and stained with Fairbanks A (25% Isopropanol, 10% Acetic 21. Kayed, R. et al. Common structure of soluble amyloid oligomers implies Acid, 0.05% Coomassie R) and destained with Fairbanks D (10% Acetic Acid) common mechanism of pathogenesis. Science 300, 486–489 (2003). afterwards. 22. Sikkink, L. A. & Ramirez-Alvarado, M. Cytotoxicity of amyloidogenic immunoglobulin light chains in cell culture. Cell Death Dis. 1, 1 (2010). Transmission electron microscopy (TEM). TEM samples were prepared by 23. Morgan, G. J. & Kelly, J. W. The kinetic stability of a full-length antibody light pipetting 10 µL samples onto a 200-mesh activated copper grid and incubating for chain dimer determines whether endoproteolysis can release amyloidogenic 1 min. The samples were washed with 2 × 10 µL H O and negatively stained with variable domains. J. Mol. Biol. 428, 4280–4297 (2016). 8 µl of a 1.5% (w/v) uranyl acetate solution for 1 min. Excess solutions were 24. Yamamoto, S. et al. Low concentrations of sodium dodecyl sulfate induce the removed with a filter paper. TEM micrographs were recorded on a JEOL JEM-1400 extension of β2-microglobulin-related amyloid fibrils at a neutral pH. Plus transmission electron microscope (JEOL Germany GmbH, Freising, Ger- Biochemistry 43, 11075–11082 (2004). many) at 120 kV. 25. Goto, Y. & Hamaguchi, K. Light role of the intrachain of the constant disulfide fragment bond in the conformation and stability of the immunoglobulin by measuring Intact CL was more resistant than reduced CL to guanidine In a Statistics and reproducibility. All representative TEM micrographs in this pub- previous paper (1) we reported on the refolding of B. J. Biochem. 86, lication have been chosen from 3 to 20 images taken from a negative stained 200- 1433–1441 (1979). mesh activated copper grid. 26. Bolotina, I. A., Chekhov, V. O. & Lugauskas, V. Y. Determination of the secondary structures of proteins by circular dichroism spectra. Calculation of Reporting summary. Further information on research design is available in the Nature the protein basic circular dichroism spectra for antiparallel and parallel β‐ Research Reporting Summary linked to this article. structures and β‐bends. Int. J. Quantum Chem. 16, 819–824 (1979). 27. Sahoo, B. R. et al. A cationic polymethacrylate-copolymer acts as an agonist for β-amyloid and an antagonist for amylin fibrillation. Chem. Sci. 10, Data availability 3976–3986 (2019). All data generated during this study are available in the article and supplementary 28. Micsonai, A. et al. BeStSel: a web server for accurate protein secondary information or from the corresponding author. Source data are provided with this paper. structure prediction and fold recognition from the circular dichroism spectra. Nucleic Acids Res. 46, W315–W322 (2018). Received: 18 December 2020; Accepted: 25 October 2021; 29. Zandomeneghi, G., Krebs, M. R. H., McCammon, M. G. & Fändrich, M. FTIR reveals structural differences between native beta-sheet proteins and amyloid fibrils. Protein Sci. 13, 3314–3321 (2004). 30. Goormaghtigh, E., Ruysschaert, J. M. & Raussens, V. Evaluation of the information content in infrared spectra for protein secondary structure References determination. Biophys. J. 90, 2946–2957 (2006). 1. Quock, T. P., Yan, T., Chang, E., Guthrie, S. & Broder, M. S. Epidemiology of 31. De Meutter, J. & Goormaghtigh, E. Evaluation of protein secondary structure AL amyloidosis: a real-world study using US claims data. Blood Adv. 2, from FTIR spectra improved after partial deuteration. Eur. Biophys. J. 50, 1046–1053 (2018). 613–628 (2021). 2. Baden, E. M., Sikkink, L. A. & Ramirez-Alvarado, M. Light chain amyloidosis 32. Horowitz, P. M. & Criscimagna, N. L. Differential binding of the fluorescent —-current findings and future prospects. Curr. Protein Pept. Sci. 10, 500–508 probe 8-anilinonaphthalene-2-sulfonic acid to rhodanese catalytic (2009). intermediates. Biochemistry 24, 2587–2593 (1985). 3. Gertz, M. Immunoglobulin light chain amyloidosis: 2020 update on diagnosis, 33. Chattopadhyay, A. & Haldar, S. Dynamic insight into protein structure prognosis, and treatment. Am. J. Hematol. 95, 848–860 (2020). utilizing red edge excitation shift. Acc. Chem. Res. 47,12–19 (2014). 4. Baden, E. M., Randles, E. G., Aboagye, A. K., Thompson, J. R. & Ramirez- 34. Catici, D. A. M., Amos, H. E., Yang, Y., van den Elsen, J. M. H. & Pudney, C. Alvarado, M. Structural insights into the role of mutations in amyloidogenesis. R. The red edge excitation shift phenomenon can be used to unmask protein J. Biol. Chem. 283, 30950–30956 (2008). structural ensembles: implications for NEMO-ubiquitin interactions. FEBS J. 5. Hurle, M. R., Helms, L. R., Li, L., Chan, W. & Wetzel, R. A role for 283, 2272–2284 (2016). destabilizing amino acid replacements in light-chain amyloidosis. Proc. Natl 35. Serrano, A. L., Lomont, J. P., Tu, L.-H., Raleigh, D. P. & Zanni, M. T. A free Acad. Sci. USA 91, 5446–5450 (1994). energy barrier caused by the refolding of an oligomeric intermediate controls 6. Kazman, P. et al. Fatal amyloid formation in a patient’s antibody light chain is the lag time of amyloid formation by hIAPP. J. Am. Chem. Soc. 129, caused by a single point mutation. Elife 9,1–23 (2020). 16748–16759 (2017). 7. Ramirez-Alvarado, M. Amyloid formation in light chain amyloidosis. Curr. 36. Martin, D. J. & Ramirez-Alvarado, M. Comparison of amyloid fibril formation Top. Med. Chem. 12, 2523–2533 (2012). by two closely related immunoglobulin light chain variable domains. Amyloid 8. Buxbaum, J. Aberrant immunoglobulin synthesis in light chain amyloidosis. J. 17, 129–136 (2010). Clin. Invest. 78, 798–806 (1986). 37. Blancas-Mejía, L. M. et al. Thermodynamic and fibril formation studies of full 9. Gallo, G., Wisniewski, T. & Ghiso, J. Short communication fibrillogenesis. Am. length immunoglobulin light chain AL-09 and its germline protein using scan J. Pathol. 145, 526–530 (1994). rate dependent thermal unfolding. Biophys. Chem. 207,13–20 (2015). NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26845-0 38. Weber, B. et al. The antibody light-chain linker regulates domain orientation 61. Phillips, S. R., Wilson, L. J. & Borkman, R. F. Acrylamide and iodide and amyloidogenicity. J. Mol. Biol. 430, 4925–4940 (2018). fluorescence quenching as a structural probe of tryptophan microenvironment 39. Karamanos, T. K. et al. Structural mapping of oligomeric intermediates in an in bovine Lens crystallins. Curr. Eye Res. 5, 611–620 (1986). amyloid assembly pathway. Elife 8,1–32 (2019). 62. Tallmadge, D. H., Huebnert, J. S. & Borkman, R. F. Acrylamide quenching of 40. Souillac, P. O. et al. Elucidation of the molecular mechanism during the early tryptophan photochemistry and photophysics. Photochem. Photobiol. 49, events in immunoglobulin light chain amyloid fibrillation: evidence for an off- 381–386 (1989). pathway oligomer at acidic pH. J. Biol. Chem. 277, 12666–12679 (2002). 41. Zhijie, Q., Dongmei, H., Min, Z. & Fink, A.L.Structuralcharacterizationofthe Acknowledgements partially folded intermediates of an immunoglobulin light chain leading to amyloid This work was supported by a grant from the DFG to J.B. (BU836/14-1) in the context of fibrillation and amorphous aggregation. Biochemistry 46, 3521–3531 (2007). the research unit FOR 2969. We thank Dragana Catici for help with REES analyses and 42. Oberti, L. et al. Concurrent structural and biophysical traits link with Laura Meier for her help with protein expression and purification. immunoglobulin light chains amyloid propensity. Sci. Rep. 7,1–11 (2017). 43. Crespo, R., Rocha, F. A., Damas, A. M. & Martins, P. M. A generic crystallization-like model that describes the kinetics of amyloid fibril Author contributions formation. J. Biol. Chem. 287, 30585–30594 (2012). P.K. and J.B. designed the study, P.K., H.E. and R.A. performed the experiments, P.K., 44. Chatani, E. & Yamamoto, N. Recent progress on understanding the H.E. and R.A. analyzed the results, P.K., R.A. and J.B. wrote the manuscript. mechanisms of amyloid nucleation. Biophys. Rev. 10, 527–534 (2018). 45. Feige, M. J., Hendershot, L. M. & Buchner, J. How antibodies fold. Trends Funding Biochem. Sci. 35, 189–198 (2010). Open Access funding enabled and organized by Projekt DEAL. 46. Huber, R., Deisenhofer, J., Colman, P. M., Matsushima, M. & Palm, W. Crystallographic structure studies of an IgG molecule and an Fc fragment. Nature 264, 415–420 (1976). Competing interests 47. Eisenhaber, F. & Argos, P. Hydrophobic regions on protein surfaces: The authors declare no competing interests. definition based on hydration shell structure and a quick method for their computation. Protein Eng. 9, 1121–1133 (1996). Additional information 48. Moelbert, S. & Emberly, E. Correlation between sequence hydrophobicity and Supplementary information The online version contains supplementary material surface-exposure pattern of database proteins. Protein Sci. 13, 752–762 (2004). available at https://doi.org/10.1038/s41467-021-26845-0. 49. Young, L., Jernigan, R. L. & Covell, D. G. A role for surface hydrophobicity. Protein Sci. 3, 717–729 (1994). Correspondence and requests for materials should be addressed to Johannes Buchner. 50. Hora, M. et al. MAK33 antibody light chain amyloid fibrils are similar to oligomeric precursors. PLoS ONE 12,1–14 (2017). Peer review information Nature Communications thanks Gareth J Morgan, Marina 51. Dear, A. J. et al. Kinetic diversity of amyloid oligomers. Proc. Natl Acad. Sci. Ramirez-Alvarado and the other, anonymous, reviewer(s) for their contribution to the USA 117, 12087 (2020). peer review of this work. 52. Arden, B. G. et al. Measuring the energy barrier of the structural change that initiates amyloid formation. Anal. Chem. 92, 4731 (2020). Reprints and permission information is available at http://www.nature.com/reprints 53. Sakurai, K. & Tomiyama, R. Enhanced accessibility and hydrophobicity of amyloidogenic intermediates of the β2-microglobulin D76N mutant revealed Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in by high-pressure experiments. J. Biol. Chem. 296, 100333 (2021). published maps and institutional affiliations. 54. Corazza, A. et al. Properties of some variants of human β2-microglobulin and amyloidogenesis. J. Biol. Chem. 279, 9176–9189 (2004). 55. Taricska, N. et al. The route from the folded to the amyloid state: exploring the potential energy surface of a drug‐like miniprotein. Chemistry 26, 1968 (2020). Open Access This article is licensed under a Creative Commons 56. Klein, W. L., Stine, W. B. & Teplow, D. B. Small assemblies of unmodified Attribution 4.0 International License, which permits use, sharing, amyloid-protein are the proximate neurotoxin in Alzheimer’s disease. adaptation, distribution and reproduction in any medium or format, as long as you give Neurobiol. Aging 25, 569–580 (2004). appropriate credit to the original author(s) and the source, provide a link to the Creative 57. Kihara, M. et al. Seeding-dependent maturation of β2-microglobulin amyloid Commons license, and indicate if changes were made. The images or other third party fibrils at neutral pH. J. Biol. Chem. 280, 12012–12018 (2005). material in this article are included in the article’s Creative Commons license, unless 58. Nokwe, C. N. et al. The antibody light-chain linker is important for domain indicated otherwise in a credit line to the material. If material is not included in the stability and amyloid formation. J. Mol. Biol. 427, 3572–3586 (2015). article’s Creative Commons license and your intended use is not permitted by statutory 59. Brown, P. H. & Schuck, P. Macromolecular size-and-shape distributions by regulation or exceeds the permitted use, you will need to obtain permission directly from sedimentation velocity analytical ultracentrifugation. Biophys. J. 90, the copyright holder. To view a copy of this license, visit http://creativecommons.org/ 4651–4661 (2006). licenses/by/4.0/. 60. Schuck, P. Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and Lamm equation modeling. Biophys. J. 78, 1606–1619 (2000). © The Author(s) 2021 10 NATURE COMMUNICATIONS | (2021) 12:6516 | https://doi.org/10.1038/s41467-021-26845-0 | www.nature.com/naturecommunications

Journal

Nature CommunicationsSpringer Journals

Published: Nov 11, 2021

There are no references for this article.