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

Learn More →

Role of mutations and post-translational modifications in systemic AL amyloidosis studied by cryo-EM

Role of mutations and post-translational modifications in systemic AL amyloidosis studied by cryo-EM ARTICLE https://doi.org/10.1038/s41467-021-26553-9 OPEN Role of mutations and post-translational modifications in systemic AL amyloidosis studied by cryo-EM 1 1 1 1 2 Lynn Radamaker , Sara Karimi-Farsijani , Giada Andreotti , Julian Baur , Matthias Neumann , 3 3 4 1 5 2 Sarah Schreiner , Natalie Berghaus , Raoul Motika , Christian Haupt , Paul Walther , Volker Schmidt , 3 6 6 7 5,8 Stefanie Huhn , Ute Hegenbart , Stefan O. Schönland , Sebastian Wiese , Clarissa Read , 1 1 Matthias Schmidt & Marcus Fändrich Systemic AL amyloidosis is a rare disease that is caused by the misfolding of immunoglobulin light chains (LCs). Potential drivers of amyloid formation in this disease are post-translational modifications (PTMs) and the mutational changes that are inserted into the LCs by somatic hypermutation. Here we present the cryo electron microscopy (cryo-EM) structure of an ex vivo λ1-AL amyloid fibril whose deposits disrupt the ordered cardiomyocyte structure in the heart. The fibril protein contains six mutational changes compared to the germ line and three PTMs (disulfide bond, N-glycosylation and pyroglutamylation). Our data imply that the disulfide bond, glycosylation and mutational changes contribute to determining the fibril protein fold and help to generate a fibril morphology that is able to withstand proteolytic degradation inside the body. 1 2 3 Institute of Protein Biochemistry, Ulm University, 89081 Ulm, Germany. Institute of Stochastics, Ulm University, 89081 Ulm, Germany. Medical Department V, Section of Multiple Myeloma, Heidelberg University Hospital, 69120 Heidelberg, Germany. Department of Asia-Africa-Studies, Middle Eastern History and Culture, University of Hamburg, 20148 Hamburg, Germany. Central Facility for Electron Microscopy, Ulm University, 89081 6 7 Ulm, Germany. Medical Department V, Amyloidosis Center, Heidelberg University Hospital, 69120 Heidelberg, Germany. Core Unit Mass Spectrometry and Proteomics, Medical Faculty, Ulm University, 89081 Ulm, Germany. Institute of Virology, Ulm University Medical Center, 89081 Ulm, Germany. email: marcus.faendrich@uni-ulm.de NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ystemic AL amyloidosis is defined by the formation of fibrils that show a width of ~9 nm. Quantification of the fibril 1,2 amyloid fibrils by immunoglobulin light chains (LCs) . end-to-end distances and contour lengths allowed us to deter- SThese fibrils can deposit at multiple sites in the body where mine the fibril persistence length at 0.74 ± 0.08 μm and its −27 −28 2 they can lead to severe impairment of vital organ functions. bending rigidity at 3.1 × 10 ±3× 10 Nm (Fig. 1c). The Untreated patients with a prominent heart involvement show a persistence length corresponds to values reported in the 29,30 high risk of death with a median survival of only seven months literature for amyloid fibrils, which vary between 16 nm and 3,4 after their initial diagnosis . Due to their natural function as part 18.5 μm, indicating that FOR001 amyloid fibrils and their of B-cell receptors and antibodies, LCs are hypervariable proteins. deposits are structurally rigid. The fibrils in these deposits interact This variability arises from the genetic recombination of variable with the surfaces of adjacent cardiomyocytes, and these interac- (V), joining (J), and constant (C) gene segments that are encoded tions occur mainly via the fibril tips (Fig. 1d), and only rarely via in the germ line (GL), somatic hypermutation, and the junctional the fibril lateral surfaces. In some cases, we find deformations in diversity at the V/J interface in the course of V/J recombination . the plasma membrane associated with the focal contact points of Altogether, there are 63–71 functional V (34–38 V and 29–33 the fibrils (Fig. 1d). The fibrils impede the contractile function of V ), 9–10 J (5 J and 4–5J ), and 5–6 C (1 C and 4–5C ) the heart, suggesting that clearance of the amyloid may help to λ κ λ κ λ segments encoded within the GL of which IGLV6-57, IGLV3-01, restore cardiac function. While there is evidence that the patient’s IGLV2-14, and IGKV1-33 were found to be overrepresented in health condition is also defined by circulating amyloid precursors 7–10 31 systemic AL amyloidosis . These findings and the fact that in the serum , our observations underpin the view that the usage of IGLV1-44 is associated with cardiac involvement imply amyloid deposits are damaging to the patient. that the LC primary structure is a key determinant for the development of amyloidosis. Further support for this view comes from observations that Cryo-EM structure of the FOR001 AL amyloid fibril.To the mutational changes that are inserted into the LCs during investigate the structure of the extracted FOR001 AL fibrils, we somatic hypermutation affect the kinetics of amyloid fibril for- imaged them using cryo-EM (Fig. 2a). The images show that mation in vitro . The effect of these mutations has frequently approximately 75% of the fibrils seen in the micrographs belong been attributed to alterations in the biophysical properties to one dominant morphology. This fibril morphology is defined of the natively folded, globular LCs, such as to a decreased by a width of ~9 nm, which agrees with the width measured by 12,13 thermodynamic stability , to increased conformational electron tomography (see above), and a crossover distance of 14–16 dynamics , or to local structural changes in a key region of ~55 nm, as measured from the recorded images. 3D reconstruc- 14–16 the protein . In other cases, it was suggested that the muta- tion of the fibril images yielded a 3D map of the dominant tions may affect the formation of specific folding or misfolding morphology with a resolution of 3.1 Å (Supplementary Fig. 1a, 14,17 intermediates or the stability or structure of the resulting Supplementary Table 1), while the remaining minor morpholo- 18,19 amyloid fibril . Post-translational modifications (PTMs) may gies could not be reconstructed. The rise and twist of the further modulate the effects of the inserted mutations. Several reconstructed fibril are 4.76 Å and −1.46°, the pitch is 117 nm studies have shown that AL fibril proteins can contain PTMs like (Supplementary Table 1). The fibril consists of a single proto- 20–23 disulfide bonds , N-terminal pyroglutamate (Pyro-Glu) filament (C1 symmetry, Fig. 2b, c). The fitted molecular model 24,25 26,27 modifications , or glycosylations . Glycosylation was found (Fig. 2d) has a model resolution of 3.1 Å, (Supplementary to be overrepresented in AL patients, indicating that this PTM Table 1). Projections of its density onto the y–z plane correspond 26,28 contributes to the pathogenesis in AL amyloidosis . Yet, the well to the two-dimensional (2D) class averages (Supplementary mechanism by which glycosylation affects fibril formation or AL Fig. 1b). The handedness of the fibrils in this sample was deter- pathogenesis has so far remained unclear. mined by platinum side-shadowing, which showed a left-handed To shed light on the structural effects of PTMs and mutational twist (Supplementary Fig. 2). The ordered core of the fibril changes on the fibril state, we have determined the structure of an consists of two segments that extend from Ser9 to Thr52 and AL amyloid fibril with cryoelectron microscopy (cryo-EM), which Ser68 to Thr108 (Fig. 2d). For residues Gln1–Pro8, Asp53–Lys67, is partially pyroglutamylated, N-glycosylated and modified by an and Val109–Ser118, which are present in the fibril as shown by intramolecular disulfide bond. The observed fibril structure is mass spectrometry (MS) (see below), no well-defined density different from previously described, nonglycosylated AL amyloid could be discerned in our map, suggesting that they are struc- 21–23 fibrils , consistent with the patient-specific nature of this turally disordered. The fibril protein contains an intramolecular disease. The mutational changes are clustered into two topolo- disulfide that connects residues Cys22 and Cys89 (Fig. 2d). gical regions of natively folded variable LC domains, but dis- persed throughout the fold of the fibril protein with no obvious structural preference. Chemical interactions stabilizing the fibril protein fold.Despite being dominated by β-sheet conformation, the fibril protein structure is profoundly different from the structure of a natively folded LC Results (Fig. 3a) and encompasses eleven parallel cross-β-sheets (β1–β11) FOR001 amyloid fibrils are structurally rigid and disrupt the that are formed by residues Val10–Ala12, Pro14–Ser21, ordered architecture of the heart muscle. The presently analyzed Asn31–Val34, Tyr37–Gln39, Thr43–Ala44, Pro45–Glu51, fibrils were extracted from the heart tissue of patient FOR001, Thr70–Leu74, Ile76–Gly78, Tyr88–Cys89, Thr91–Glu93, and who suffered from advanced cardiac AL amyloidosis and Thr105–Leu107 (Fig. 3a, b). The β-strands in these sheets interact in underwent a heart transplantation at the age of 51 years. Analysis the direction of the fibril z axis through backbone hydrogen bonds as of sections of cardiac tissue with scanning electron microscopy well as through side-chain interactions, including the stacking of (SEM) demonstrates that the fibrils form large-sized, extracellular aromatic and polar residues (Supplementary Fig. 3a, b). The fibril amyloid deposits that infiltrate and disrupt the ordered structure protein fold is defined by buried electrostatic interactions, such as of the cardiomyocytes (Fig. 1a, b). Using scanning transmission between Glu84 and Lys46 (Supplementary Fig. 3c), and by several electron microscopy (STEM), we obtained tomograms of the buried patches of hydrophobic residues, such as the one formed by three-dimensional (3D) structure of the cardiac amyloid deposits residues Val10, Ala12, Val34, and Trp36 (Fig. 3c). The fibril qua- (Fig. 1b). The deposits are composed of haphazardly arranged ternary structure is stabilized by an interlocking of the protomers in 2 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE ab 0.15 measured data points fitted curve 0.1 0.05 0 0.1 0.2 0.3 0.4 contour length [μm] Fig. 1 Tissue deposits of FOR001 amyloid fibrils. a SEM overview of FOR001 heart tissue. The fibril deposits between cardiomyocytes are marked with blue asterisks. Scale bar: 1 μm. b STEM tomogram. Top: virtual section. The fibril deposit is marked with a blue asterisk. Bottom: rendered tomogram of the fibril deposit. Blue: amyloid fibrils. Red: membranes. Scale bar: 100 nm. c Analysis of the persistence length based on a fit of the plot showing the squared end-to-end distance versus the contour length using Eq. (1). The blue symbols show the measured data points (n = 195) and the red line the fit. d Region of the tomogram, showing the interactions of the fibrils (blue) with the cardiomyocyte membrane (red). Scale bar 100 nm. the direction of the z-axis that is caused by a 14-Å height change of immunoglobulin LCs and their GL segments (see “Methods” for the fibril protein backbone (Supplementary Fig. 3d). details) revealed that the FOR001 fibril protein stems from the GL segment IGLV1-51*02. The GL precursors of the J and C seg- ments could not be identified unambiguously. In case of the C Primary structure of the FOR001 fibril protein. The FOR001 segment, the number of residues (8, Gly111–Ser118) was too fibril protein sequence was determined with electrospray- small to allow any GL assignment. In case of the J segment, ionization MS (Supplementary Fig. 4). DNA sequencing of IGLJ2*01, IGLJ3*01 (which are identical) and IGLJ3*02 are all bone-marrow-derived cDNA was also attempted but failed to possible precursors (IGLJ3*02 differs from IGLJ2*01 and produce a sequence that matched the sequence obtained by direct IGLJ3*01 only in the first residue, which is mutated in the sequencing of the fibril protein. Three major fibril protein species FOR001 protein sequence), also precluding a unique GL assign- were revealed by MS (Supplementary Fig. 5a). Two of these could ment. The FOR001 sequence shows six mutational changes be assigned to the LC fragments Pyro–Glu1 to Ser118 and Ser2 to compared with the GL segments IGLV1-51*02, IGLJ2*01, Ser118 (Supplementary Fig. 5b), both containing a disulfide bond. IGLJ3*01, and IGLJ3*02: Lys17Asn, Asn52Thr, Asn53Asp, The third species corresponds to the fragment containing residues Gly82Ala, and Asp93Glu in the V segment and Xaa99Gly in the J Val3–Ser118 and a disulfide bond, although alternative mass segment (Fig. 4a, b, indigo). The mutation Lys17Asn inserts the assignments are also possible for the recorded MS peak (Sup- glycosylation site. In addition, residues Leu97–Ala98 form the plementary Fig. 5b). These analyses reveal two PTMs: a disulfide variable junction between the V and J segments that arises from bond and a pyroglutamylation (Supplementary Fig. 5c) that genetic V/J recombination (Fig. 4a, b, green). affects only a fraction of the fibril proteins. The third PTM of the fibril protein is an ~2-kDa N-glycosylation, which is demon- strated by an electrophoretic band shift of the refolded FOR001 The location of the mutations in known fibril structures and in fibril protein upon addition of N-glycosidase but not with natively folded V domains. To identify the possible role of O-glycosidase (Supplementary Fig. 5d). The carbohydrate can be mutations in LC aggregation, we analyzed their position within seen as extra density within our 3D map that could not be the known fibril structures and natively folded LCs. We find that assigned to the polypeptide chain (Fig. 2b, d, red star). It pro- the majority of the mutations occur at solvent-exposed positions trudes from residue Asn17, which forms part of the only cano- in the fibril, specifically in the FOR001 fibril. Five FOR001 nical N-glycosylation site (Asn–Xaa–Thr/Ser) in the fibril protein mutations (Lys17Asn, Asn52Thr, Gly82Ala, Asp93Glu, and sequence. Xaa99Gly) are part of the ordered fibril core, the sixth mutation (Asn53Asp) affects a structurally disordered region (Fig. 4a, b). Identification of the GL segments and mutational changes. One residue (position 53 in the FOR001 fibril protein) is mutated Comparisons of the FOR001 LC with known sequences of in all known AL fibril structures (Fig. 4a, b) and was identified NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 3 squared end-to-end distance [μm²] ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 Fig. 2 Cryo-EM structure of the FOR001 AL amyloid fibril. a Cryo-EM image of FOR001 amyloid fibrils. Scale bar is 100 nm. The dataset consists of 3033 micrographs. b Cross section of the map obtained by summing five central slices. c Side view of the map (left) and molecular model (right), showing the left-handed fibril twist. d Cross section of the map with the molecular model overlaid. The color coding of the model is the same in panels (c) and (d), that is, light blue refers to the N-terminal segment of the ordered fibril protein (residues Ser9–Thr52), while the deep-red segment refers to the C-terminal segment (residues Ser68–Thr108). The two segments are cross-linked through a disulfide between Cys22 and Cys89. The red star in (b) and (d) indicates the glycosylation site at Asn17. Fig. 3 Location of the secondary structural elements and mutational sites in the fibril structure. a Amino acid sequence of the FOR001 fibril protein and secondary structural elements of the FOR001 fibril protein (PDB 7NSL) and of a crystal structure of a natively folded LC (PDB 4ODH 10.2210/pdb4ODH/ pdb) containing an IGLV1-51*02 segment. Arrows indicate β-strands in the structure, rainbow-colored from N (blue) to C terminus (red). Dotted lines represent disordered segments. Red star: location of the glycosylation. b Stack of seven protein layers of the fibril, showing the β-strands β1–β11 with the same coloring as in (a). c Schematic representation of the fibril protein fold. Red star: location of the glycosylation. 4 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE Fig. 4 Location of the mutations in known AL amyloid fibrils and natively folded V domains. a Location of the mutations in known AL amyloid fibril structures. The fibrils are derived from the GL segments IGLV1-51*02 λ1 (FOR001, this study, PDB 7NSL) IGLV1-44*01 λ1 (FOR006, PDB 6IC3 10.2210/ 21 23 22 pdb6IC3/pdb) , IGLV3-19*01 λ3 (FOR005, PDB 6Z1O 10.2210/pdb6Z1O/pdb) , and IGLV6-57*02 λ6 (AL55, PDB 6HUD 10.2210/pdb6HUD/pdb) . Disordered parts of the fibril proteins are indicated by dotted lines. Black: CDRs; yellow: Cys. Indigo: mutations in the CDRs and one residue before or after a CDR; magenta: mutations in framework regions; green: residues in the junctional region at the V/J interface. Red star: location of the glycosylation. b Sequence alignment of the four fibril proteins. CDRs are marked with gray boxes. Color coding as in (a). c Location of mutations in the corresponding, natively folded LC V domains that are based on the GL segments IGLV1-51*01 (PDB 5JZ7 10.2210/pdb5JZ7/pdb), IGLV1-44*01 (PDB 6QB6 10.2210/ pdb6QB6/pdb) IGLV3-19*01 (PDB 6Q0E 10.2210/pdb6Q0E/pdb) and IGLV6-57*02 (PDB 7JVA 10.2210/pdb7JVA/pdb) CDRs are marked in black. Color coding as in (a). previously as a mutational hotspot in systemic AL amyloidosis . fibril, which does not contain an internal disordered segment, the It is located in the central disordered segments that are present in PDRFSGS motif is not preceded by a segment with an the FOR001, FOR005, and AL55 fibril structures (Fig. 4a), indi- aggregation score of 5 (Supplementary Fig. 6a, b). These data cating that this residue does not substantially affect the fibril suggest that mutations and aggregation-prone segments help to stability. About 79 ± 21% of the mutations of the four known AL define the specific fold of the observed fibril morphologies. amyloid fibrils reside within an ordered structural segment of the The mutations may additionally affect the native state, as they are V domain (Supplementary Table 2). This value is identical, found to be clustered into two regions of the globular V domains: L L within error, to the percentage of residues forming the ordered One region is formed by the complementarity-determining region fibril parts (77 ± 6%), suggesting that the mutations are not pre- (CDR) mutations around the lower-right rim of the native fold when ferentially located in the fibril core. However, mutations within it is oriented as in Fig. 4c (blue spheres). This part of the structure is the ordered part of the fibril protein are generally well- involved in forming the antigen-binding site and includes the shared accommodated in the structure and can add specific interac- mutated residue (position 53 in FOR001). The second region is tions. In the FOR006 fibril, they enable the interactions between formed by the framework mutations in the upper part of the V Arg25 (mutated from Ser) and Glu84 [21]; in the FOR001 fibril, domain (Fig. 4c, magenta spheres). Both clusters correlate with between Ala82 (mutated from Gly) and Leu48. previously described mutational hotspots of AL fibril proteins . In a next step, we analyzed the position of the aggregation- Taken together with the observations above, we conclude that the prone segments in the fibril structures. These segments were mutational changes in AL patients preferentially affect two structural 21,23 identified here based on their theoretic aggregation score . sites in the native protein state. In addition, they contribute The aggregation-prone segments (as defined by an aggregation interactions to support the formation of a specific fibril morphology. score of 5) occur within the stable core of the FOR001 fibril and other known AL amyloid fibrils (Supplementary Fig. 6a, b). By Effect of the N-glycosylation on fibril formation. The glycosy- contrast, the structurally disordered regions of these fibrils lation site is exposed on the fibril surface (Fig. 2d) and on the typically show an aggregation score of 0 (Supplementary Fig. 6b). surface of a natively folded LC (Fig. 5a, b). In the native LC, it is The fibrils with an internal disordered segment (FOR001, far away from the interface to the heavy chain (Fig. 5a) or to the FOR005, and AL55) share a conserved sequence motif second LC in a LC dimer (Fig. 5b). These observations indicate (PDRFSGS) with low aggregation propensity that is that the glycosylation does not strongly interfere with the ability N-terminally preceded by a highly aggregation-prone segment of the FOR001 LC to assemble into antibodies or LC dimers. To (aggregation score 5, Supplementary Fig. 6a, b). In the FOR006 investigate the influence of the glycosylation on fibril formation NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 a b c V 80 V V H L L C 0 10 20 30 40 50 60 70 time (h) In vitro In vitro Ex vivo 100 glycosylated deglycosylated M M M - - - - - - P Proteinase K + + + + + + + + + + + + 80 80 62 62 50 50 31 31 20 4 5 0 1 2 3 6.5 6.5 6.5 time (min) Fig. 5 Effect of glycosylation on the formation of fibrils from FOR001 fibril protein. a Ribbon representations of a fragment antigen binding that contains a LC with an IGLV1-51*02 GL segment (PDB 4ODH 10.2210/pdb4ODH/pdb). The V and the C domain, as well as the variable heavy (V ) and constant L L H heavy (C ) domains are labeled. The LC is marked green, the heavy chain is displayed in gray. Red sphere: residue homologous to the FOR001 glycosylation site. b Crystal structure of a LC dimer encompassing an IGLV1-51*02 GL segment (PDB 5MUD 10.2210/pdb5MUD/pdb). One LC in the dimer is marked green, the other in gray. Red sphere: as in (a). c Fibril-formation kinetics of refolded FOR001 fibril protein as obtained from real-time measurements of the ThT fluorescence intensity. Blue: deglycosylated; red: glycosylated protein. d Coomassie-stained denaturing-protein electrophoresis gels of samples to estimate the proteolytic stability of ex vivo FOR001 fibrils and fibrils formed in vitro from deglycosylated and glycosylated FOR001 protein. Each gel was replicated three times. M: marker. e Densitometric quantification of the fibril protein band (n = 3) of ex vivo fibrils (gray) and in vitro fibrils from deglycosylated (blue) and glycosylated FOR001 protein (red) after digestion with proteinase K for different periods of time. The band intensity of the sample before proteinase K addition was set to 100%. Based on a one-tailed Welch t-test, the amounts of glycosylated and deglycoslated fibril proteins differ from one another with a p-value of 0.032 and 0.049 at time points 0 min and 1 min, respectively. Error bars represent the standard deviation. in vitro, we purified and refolded FOR001 fibril protein from the four known ex vivo amyloid fibril structures from systemic AL heart. The glycosylation was removed from a fraction of the amyloidosis. Each of these structures is different (Supplementary refolded protein. Both the glycosylated and the deglycosylated Fig. 7). The observed differences include the precise position of protein variants were able to form fibrils in vitro, as indicated by the β-strands and disordered regions within the fibril protein time-resolved fibrillation measurements with the amyloid- sequence, the presence or absence of internal cavities, the topo- binding dye thioflavin T (ThT) (Fig. 5c). The deglycosylated logical arrangement of the secondary structural elements, the FOR001 fibril protein aggregated much faster under these con- organization of the network of chemical interactions that stabilize ditions than the glycosylated fibril protein (Fig. 5c), demon- the misfolded conformational state, and the involvement of strating that a glycosylation retards rather than accelerates fibril PTMs. However, the fibrils share structural features such as a formation in vitro. We further investigated the proteolytic sta- fibril core that is mainly derived from the LC V domain, the bility of the glycosylated and deglycosylated fibril samples presence of an intramolecular disulfide bond, and the tertiary (Fig. 5d). Both in vitro fibril samples are substantially less stable structures of the fibril proteins are substantially different from a to proteolysis with proteinase K than the ex vivo fibrils, although natively folded LC. the glycosylated in vitro fibrils seem slightly more stable to pro- In the FOR001 fibril, we identified three PTMs: a disulfide teolysis than the deglycosylated in vitro fibrils (Fig. 5e). bond (Figs. 2d and 3c), a Pyro-Glu modification that affects only a fraction of the fibril proteins (Supplementary Figs. 4 and 5a, b), and a carbohydrate moiety that is linked by N-glycosylation Discussion (Supplementary Fig. 5d). The disulfide bond corresponds to the In this study, we obtained the cryo-EM structure of a glycosylated canonical disulfide of a natively folded LC V domain . It occurs λ1 AL amyloid fibril that was extracted from the heart tissue of in all previously described cryo-EM structures of ex vivo AL patient FOR001. Adding this structure to the previously pub- 21–23 21–23 amyloid fibrils , where it connects two chain segments that lished ex vivo AL amyloid fibril structures , there are now 6 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications m molecular mass (kDa) 0 min 1min 2 min 5 min 0 min 1 min 2 min 5 min 0 min 1 min 2 min 5 min fibril protein (%) relative ThT flourescence (%) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE present an orientation that is switched by 180° relative to the fibril protein structures or fibril morphologies are associated with 21 44–46 native state . This disulfide has profound effects on the mis- different disease variants . Ex vivo fibril morphologies are folding and on the fibril structure, as it restricts the conforma- often different from fibrils formed in vitro from the same 20,23,44,47–49 tional freedom of the fibril protein and leads to the requirement proteins and more protease-stable than their in vi- 47,49 of a rotational switch around the disulfide bond as a key mole- tro formed counterparts . In particular, the latter observations cular event in the misfolding of amyloidogenic LCs . led to the hypothesis that pathogenic amyloid fibril structures 24,25 Pyro-Glu residues were previously observed in some but may have been selected inside the body by their high proteolytic 21–23 not all AL fibril proteins . They also occur in N-terminally resistance and ability to escape endogenous clearance 33 49,50 truncated Alzheimer’sAβ peptide and may accelerate the systems . The present observations further support this view fibrillation of this peptide in vitro . In the FOR001 fibril, a sig- as they imply that the disulfide formation, the glycosylation, and nificant fraction of the fibril protein lacks the Pyro-Glu mod- the mutational changes inserted during somatic hypermutation ification (Supplementary Figs. 4 and 5a, b), and the modification helped to shape the specific fibril morphology that is associated affects a disordered region of the fibril structure (Fig. 3a). While with patient FOR001 and which shows a high protease stability we cannot exclude that Pryo-Glu can affect the rate of aggrega- (Fig. 5d, e). tion, there is, based on our data, no strong evidence to suggest In summary, we have shown that in AL amyloidosis, fibril that Pyro-Glu is a major driver for the generation of the specific deposits can interfere with the physiological function of heart fibril morphology observed in patient FOR001. tissue by disrupting its ordered architecture. PTMs and the spe- The third PTM is an N-glycosylation. The Asn–Xaa–Ser/Thr cific mutational changes that characterize amyloidogenic LCs N-glycosylation motif occurs more frequently in AL LC sequen- facilitate the formation of patient-specific fibril morphologies that 35,36 ces than in normal LCs , and glycosylated LCs are over- are able to survive under the otherwise hostile and proteolytic represented in systemic AL amyloidosis patients , which conditions inside the body. The resulting fibrils are thus able to indicates that LC glycosylation is important for systemic AL accumulate, proliferate, and cause damage to the surrounding amyloidosis. However, the biophysical basis of this effect tissue. The differences in the available cryo-EM structures of remained obscure. Several studies demonstrated that glycosyla- fibrils from AL amyloidosis patients underline the patient-specific 37,38 tion can stabilize the native fold of globular proteins , reduce nature of the disease and make clear the need for investigating AL the protein conformational dynamics , or increase the solubility fibril structures from many patient cases in order to identify 37,38 of proteins , indicating that a glycosylation should render a common principles of AL amyloid formation. protein less amyloidogenic. Indeed, deglycosylated FOR001 fibril protein aggregates faster than glycosylated protein, as demon- Methods strated here by in vitro fibrillation measurements with ThT Source of the fibril-containing tissue. The fibrils were extracted from the (Fig. 5c). explanted heart of a male patient (FOR001), who suffered from systemic AL amyloidosis with cardiac involvement and underwent cardiac surgery at the age of A potentially more important effect of glycosylation could be 51 years. The underlying condition was a monoclonal gammopathy. The patient that it masks the amyloid deposits within the tissue, preventing was treated within the heart-transplant program of the University Hospital Hei- their clearance by body-own mechanisms . A similar amyloid- delberg. The explanted heart tissue was stored at −80 °C. The study was approved masking effect is known for serum amyloid P component that is by the ethical committees of the University of Heidelberg (123/2006) and of Ulm University (203/18). Informed consent was obtained from the patient for the also glycosylated. The carbohydrate of the FOR001 protein is analysis of the amyloid deposits. located on the surface of the fibril structure (Fig. 2b, d) and of the natively folded LC, as indicated by analysis of the crystal structure Visualization of amyloid fibril deposits in heart tissue using STEM and SEM. of a homologous LC (Fig. 5a, b). Hence, the carbohydrate covers Frozen pieces of FOR001 heart tissue (~1 mm )were fixed in a solution of 0.1% (w/v) part of the surface of the fibrils or its biological precursors, glutaraldehyde, 4% (w/v) paraformaldehyde, and 1% (w/v) saccharose in 0.1 M sodium consistent with a protective effect and with observations that phosphate buffer (pH 7.3) overnight at 4 °C. The further sample preparation for STEM glycosylated in vitro fibrils from FOR001 fibril protein might be and SEM was based on a protocol from a previous publication .In brief,the fixed tissue pieces were cut into ~200-μm slices with a scalpel and then high-pressure frozen slightly more stable to proteinase K digestion than in vitro formed with a Compact 01 high-pressure freezing device (Engineering Office M. Wohlwend) fibrils from deglycosylated protein. However, both in vitro fibrils and freeze-substituted in a medium consisting of 0.1% (w/v) uranyl acetate, 0.2% (w/v) are much less stable to proteinase K digestion than the ex vivo osmium tetroxide, and 5% (v/v) water in acetone using an AFS2 freeze-substitution FOR001 fibrils (Fig. 5d, e). device (Leica Microsystems) with which the temperature was raised from −90 °C to room temperature over a period of 19 h. Afterward, the tissue pieces were embedded in The most profound effect of glycosylation suggested by our epoxy resin (Sigma-Aldrich) starting with a mixture of 30% (v/v) resin in acetone for data is that it helps to define the fibril protein fold in this patient. 1 h followed by 60% (v/v) resin in acetone for 3 h and 100% resin overnight and The fibril protein tertiary structure in the glycosylated λ1 fibril of polymerization in fresh 100% resin by incubation at 60 °C for 48 h. For SEM imaging, FOR001 is significantly different from a previously described, 200-nm thin sections were cut from the polymerized samples using the ultra- microtome Ultracut (Leica Microsystems) equipped with a 45° diamond knife (Dia- nonglycosylated λ1 fibril . The previous structure, and all known tome). Sections were mounted on glow-discharged silicon wafers and stained with a structures of nonglycosylated AL amyloid fibril, is defined by a 0.3% (w/v) lead citrate solution in water for 1 min, washed with distilled water, dried, juxtaposition of the N- and C-terminal ends of the fibril protein and imaged in a Hitachi S-5200 field emission scanning electron microscope, detecting (Supplementary Fig. 7). In the FOR001 fibril, which carries the the secondary electron signal at 5 kV in analysis mode. For STEM, 300-nm thin glycosylation in the N-terminal end of the fibril protein, no such sections were cut with a 45° diamond knife and processed in a similar way as described in a previous publication . In brief, sections were mounted on copper grids with 200 juxtaposition is observed and the N- and C-terminal ends of the parallel grid bars (Plano). After attachment of 15-nm colloidal gold fiducials (Aurion), polypeptide chain are spread apart in the tertiary structure. sections were coated with carbon by electron-beam evaporation in a Baf 300 (BalTec). Hence, the glycosylation may help to favor the FOR001 fibril Tomograms were acquired with a STEM JEM-2100F (JEOL) operated at 200 kV. Tilt protein fold by preventing the association of the N- and series were acquired from −72° to +72° with a 1.5° increment using the bright-field detector. The pixel size was 1.395 nm. Tilt series were reconstructed to tomograms by C-terminal ends of the fibril protein. weighted back projection using an emulated simultaneous iterative-reconstruction Several previous observations have suggested that the fibril technique-like filter (20 iterations) and segmentation of fibrils and cell membranes was protein fold and the fibril morphology are crucial for the 53 performed with the IMOD software package , version 4.9.0. pathological process of an amyloid disease. Consistent fibril protein folds and fibril morphologies can be found in different Measurement of the persistence length. A total of 197 amyloid fibrils in heart patients or animals that are affected by the same disease variant tissue were traced in the STEM tomograms and were then available as polygonal 42–44 and allelic variant of the fibril precursor protein . Different chains. Based on this representation, the squared end-to-end distance R, i.e., the NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 Cartesian plane distance between the starting and end points, and the contour particles were subjected to a 3D classification with the initial model as a reference. length L, i.e., the sum of the length of all line segments of the polygonal chain, were This treatment resulted in a map with fibril-like features. Using this map as a computed for each fibril. The persistence length P of the fibrils was then deter- reference, the particle set selected from 2D classification was subjected to several mined by regression analysis using the formula rounds of 3D classification, 3D refinement, and post-processing. The resolution of the resulting map was estimated to be 3.6 Å. To obtain a more homogeneous set of P L P particles, fibrils were picked more selectively, yielding 43,308 particles with a box R ¼ 2PL 1  1  e ð1Þ size of ~270 Å. Using the previously obtained post-processed map as a reference 54 (with adjusted box size), subsequent steps of 3D classification, 3D refinement, and according to Kollmer et al. and using the curve-fitting tool in Matlab (MATLAB post-processing yielded a map at a resolution of 3.4 Å. All manually picked par- 2019, The MathWorks). P is contained in the confidence interval from 0.66 μmto ticles were retained in this reconstruction. Further improvement of map resolution 0.81 μm in 95% of all cases. The bending rigidity B is computed based on P was accomplished through Bayesian polishing, resulting in a final map resolution according to the formula of 3.1 Å, based on the value of the FSC curve for two independently refined half- B ¼ k TP ð2Þ maps at 0.143. An estimated map-sharpening B-factor of −67.383 Å was applied. The map had a twist of −1.45566° and a rise of 4.76311 Å. These values agree with where k and T denote the Boltzmann constant and the temperature (300 K), the twist value calculated from the measured crossover distances on the cryo-EM 21,55 respectively . micrographs, and with the rise value measured from the micrograph power spectra. Fibril extraction from FOR001 heart tissue. Fibrils were extracted from FOR001 56 Model building and refinement. The software Coot v0.8.9 was used to manually heart tissue using a previously established protocol . In brief, 250 mg of heart build the protein model de novo. First, the 3D map was traced and a poly-L-Ala tissue was diced with a scalpel and washed five times with Tris Calcium Buffer chain created representing the protein backbone. The residues of this chain were (TCB) [20 mM Tris, 138 mM NaCl, 2 mM CaCl , and 0.1% (w/v) NaN , pH 8.0]. 2 3 then mutated to the FOR001 fibril protein sequence as determined by MS. This Each washing step consisted of the addition of 0.5 mL of ice-cold TCB to the pellet, initial model was then further improved by subsequent rounds of manual and the homogenization of the tissue with a Kontes Pellet Pestle, and the separation of automated refinement (phenix.real_space_refine) as implemented in Phenix the tissue from the supernatant by centrifugation for 5 min at 3100 × g and 4 °C, v1.16 . Noncrystallographic symmetry and secondary-structure restraints were followed by the removal of the supernatant. The tissue pellet after the fifth washing imposed. Model-based automated sharpening of the map (phenix.auto_sharpen) step was resuspended in 1 mL of TCB containing 5 mg/mL Clostridium histolyti- yielded an improved map, which was used to further refine the model. The quality cum collagenase (Sigma-Aldrich) and one tablet of complete ethylenediaminete- of the model was assessed using the MolProbity -generated validation report. traacetic acid (EDTA)-free protease-inhibitor cocktail (Roche) per 7 mL of TCB. Modeling parameters are listed in Supplementary Table 1. After an overnight incubation of the digest at 37 °C in a horizontal orbital shaker (750 rpm), the sample was centrifuged for 30 min at 3100 × g and 4 °C. The pellet was subjected to ten washing steps which were performed in the same manner as Protein sequence determination by electrospray-ionization MS. About 2 μgof the TCB washing steps described above, except that 0.5 mL of Tris EDTA buffer refolded, lyophilized and glycosylated fibril protein was resuspended in 15 μLof [20 mM Tris, 140 mM NaCl, 10 mM EDTA, and 0.1% (w/v) NaN , pH 8.0] was buffer [280 mM Tris/HCl pH 6.8, 9% (w/v) sodium dodecyl sulfate (SDS), 33.3% used instead of TCB. The pellet from the last wash was resuspended in 0.5 mL of (w/v) glycerol, and 100 mM dithiothreitol] and processed by denaturing protein gel ice-cold distilled water, mixed with a pipette, and centrifuged for 5 min at 3100 × g electrophoresis. Afterward, the gel band of the fibril protein was cut out and and 4 °C. The fibril-containing supernatant was retained and the pellet was sub- washed by a 10-min incubation in the respective protease buffer (see below), and mitted to nine more cycles of resuspension in water and centrifugation. The subsequently, in a mixture of 50% (v/v) protease buffer and 50% (v/v) ACN for supernatants were retained to check for the presence of fibrils. 10 min. These incubation steps were repeated twice, followed by vacuum drying. Dried gel slices were reduced with 5 mM dithiothreitol (AppliChem) in 50 mM ammonium bicarbonate, pH 8.0, for 20 min at room temperature and subsequently Platinum side-shadowing and TEM. The handedness of the fibrils was deter- alkylated with 55 mM iodoacetamide (Sigma-Aldrich) in 10 mM ammonium mined by platinum side-shadowing and TEM. Formvar and carbon-coated 200 bicarbonate for 20 min at 37 °C. The gel slices were placed in five different protease mesh copper grids (Plano) were glow discharged for 40 s at 40 mA using a PELCO solutions (trypsin in 50 mM ammonium bicarbonate, pH 8.0; LysC in 50 mM easiGlow glow-discharge cleaning system (Ted Pella). About 15 μL of the fibril ammonium bicarbonate, pH 8.0; elastase in 50 mM Tris/HCl, pH 9.0; chymo- solution were applied to the grid and incubated for 30 s at room temperature. The grid was blotted using filter paper (Whatman) to remove excess fluid. The grid was trypsin in 50 mM Tris/HCl buffer, pH 8.0, 10 mM CaCl ; pepsin 40 mM HCl, pH 1.5). Each protease was used at 0.33 ng/μL concentration and digestion was carried washed three times with 10 μL of distilled water and dried at room temperature. out overnight at 37 °C (except for chymotrypsin at 25 °C). The resulting peptides Platinum was evaporated at an angle of 30° onto the grid to form a 1-nm-thick were released from the gel slices in two steps: the first step was to add 20 μLof a layer by use of a Balzers BAF 300 coating device. Grids were imaged using a JEM- solution containing 50% (v/v) ACN and 0.1% (v/v) TFA; the second step was an 1400 TEM (JEOL) that was operated at an acceleration voltage of 120 kV. The incubation in an ultrasonic bath (Bandelin Sonorex Super 10 P) at 100% intensity images were recorded with an F216 camera (TVIPS). for 10 min each. ACN was evaporated and samples were filled to 15 μL with 0.1% TFA (v/v). Cryo-EM sample preparation and data collection.C-flat 1.2/1.3 400-mesh holey Samples were separated by liquid chromatography using a U3000 RSLCnano carbon-coated grids (Science Services) were glow-discharged at 40 mA for 40 s (Thermo Fisher Scientific) online coupled to the mass spectrometer with an using a PELCO easiGlow glow-discharge cleaning system (Ted Pella). Conditions Acclaim PepMap analytical column (75 μm × 500 mm, 2 μm, 100-Å pore size, of grid preparation were optimized with the help of a Vitrobot Mark 3 (Thermo Thermo Fisher Scientific) in combination with a C18 μ-precolumn Fisher Scientific) and checked in a 200-kV JEM 2100 F transmission electron (0.3 mm × 5 mm, PepMap, Dionex LC Packings, Thermo Fisher Scientific). First, microscope (JEOL) that was equipped with a DE12 detector (Direct Electron). The samples were washed with 0.1% (v/v) TFA for 5 min at a flow rate of 30 μL/min. grids for data collection were prepared by application of 3.5 μLof fibril solution to Subsequent separation was carried out employing a flow rate of 250 nL/min using a a grid, incubation for 30 s at >95% humidity, both-side blotting using filter paper gradient consisting of solvent A [0.1% (v/v) formic acid] and solvent B [86% (v/v) (Whatman), and plunging into liquid ethane (~103 K). The data set was recorded ACN, 0.1% (v/v) formic acid]. The main column was initially equilibrated in a with a Titan Krios transmission electron microscope (Thermo Fisher Scientific) at mixture containing 5% (v/v) solvent B and 95% (v/v) solvent A. For elution, the 300 kV and applying a Gatan imaging filter with a 20-eV slit. The images were percentage of solvent B was raised from 5 to 15% over a period of 10 min, followed recorded with a K2-Summit detector (Gatan) in counting mode. The software by an increase from 15 to 40% over 20 min. Fractions from the main column SerialEM v3.7 was used for data collection. In total, 3033 micrographs were col- directly eluted into the ionization module and were further analyzed by MS. lected from a single grid. See Supplementary Table 1 for further details. Global Samples were measured using an LTQ Orbitrap Velos Pro system (Thermo fibril parameters, such as width and crossover distance, were measured using Fiji Fisher Scientific). The mass spectrometer was equipped with a nanoelectrospray v1.52 . The proportion of fibrils showing the reconstructed morphology was ion source and distal-coated SilicaTips (FS360-20-10-D, New Objective). The determined by analyzing all fibrils (length at least 200 nm) in 100 micrographs. instrument was externally calibrated using standard compounds (LTQ Velos ESI Positive Ion Calibration Solution, Pierce, Thermo Scientific). The system was Reconstruction of the 3D map. Motion correction and dose-weighting was car- operated using the following parameters: spray voltage, 1.5 kV; capillary ried out with MotionCor2 . For predicting, refining, and correction of the con- temperature, 250 °C; S-lens radio-frequency level, 68.9%. The software XCalibur trast transfer function, Gctf v1.06 was used. Helical reconstruction was 2.2 SP1.48 (Thermo Fisher Scientific) was used for data-dependent MS/MS performed using Relion 3.0 .As a first step, 279,338 particles were picked analyses. Full scans ranging from mass-to-charge ratio (m/z) 370–1700 were manually with a box size of ~312 Å and an interbox distance of 34.6 Å (~11%). acquired in the Orbitrap at a resolution of 30,000 (at m/z 400) with automatic gain After a reference-free 2D classification using 279 classes and a regularization value control enabled and set to 10 ions and a maximum fill time of 500 ms. Collision- of T = 2 in the first run, another 2D classification was performed with 50 classes. induced dissociation was employed as the fragmentation method on individual 2D classes were selected based on the visibility of a z-axial repeat at ~4.8 Å and bad sample sets. Per survey scan, 10 ions were selected. Single charged ions were classes were excluded. An initial 3D model was generated using Relion’s initial- rejected and the m/z peaks of fragmented single-charged ions were excluded from model job. To generate a better reference for 3D classification, fibrils with visible fragmentation for 60 s. In the linear ion trap, the automatic gain control was set to crossovers were picked from 28 micrographs, resulting in 244 particles. These 10,000 ions and a maximum fill time of 100 ms. For MS/MS fragmentation, a 8 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE normalized collision energy of 35% with an ‘activation q’ of 0.25 and an activation [500 mM sodium phosphate, pH 7.5] (New England Biolabs), 1/2 V water, and Prot time of 30 ms was used. The resulting fragments were analyzed using the linear 1/10 V PNGase F in 20 mM Tris/HCl, 50 mM NaCl, and 5 mM EDTA, pH 7.5 Prot ion-trap part at rapid scan speeds. Subsequent detection of fragmentation spectra (New England Biolabs), and incubated for 1 h at 37 °C. For O-deglycosylation, the was performed in the Orbitrap mass analyzer at a resolution of 7500. heated sample was mixed with 1/5 V of a 10% (v/v) Nonident P40 solution Prot For de novo sequencing, the Peaks AB Software (Bioinformatics Solutions) was (New England Biolabs), 1/5 V 10 × Glycobuffer 2 (New England Biolabs), 3/10 Prot used. The resulting sequence was then used as a target for further analyses using V water, 1/10 V O-glycosidase in 20 mM Tris/HCl, 50 mM NaCl, and 1 mM Prot Prot the Peaks X software suite (Bioinformatics Solutions) in order to confirm the de EDTA, pH 7.5 (New England Biolabs), and 1/5 V neuraminidase in 20 mM Prot novo sequence. For all analyses, the mass accuracy was set to 10 ppm on intact Tris/HCl, 50 mM NaCl, and 5 mM EDTA, pH 7.5 (New England Biolabs), and peptide masses and 0.5 Da. Various PTMs were considered, including the incubated for 10 min at 37 °C. The deglycosylation was checked by denaturing deamidation of Asn or Gln residues, pyroglutamate modifications (Gln), oxidation protein-gel electrophoresis. N- and O-glycosylated fetuin protein, which we pur- of Met, as well as carbamidomethylated Cys as a result of the alkylation. Ile/Leu as chased as a 10 mg/mL solution from New England Biolabs, was used as a control well as Gln/Lys have the same molecular weights and could not be uniquely substance. For the experiments reported in Fig. 5, the lyophilized, refolded FOR001 determined. These residues were assigned based on homology considerations. protein was dissolved in water at 1 mg/mL concentration and denatured—without SDS and dithiothreitol—for 10 min at 100 °C, as the SDS in the denaturing buffer was found to interfere with the subsequent reversed-phase chromatography. MS analysis of the total fibril protein mass. Lyophilized FOR001 fibril protein was resuspended in Glycoprotein Denaturing Buffer (New England Biolabs) and Protein-concentration measurement based on the intrinsic absorbance at subsequently it was deglycosylated without the heating step of the standard pro- 280 nm. About 40 μL of protein solution was mixed with 160 μL of 7.5 M gua- tocol. The deglycosylated protein was diluted with 0.1% TFA (v/v) to obtain a nidine hydrochloride (Carl Roth) in 25 mM sodium phosphate buffer, pH 6.5. The concentration of 66 μg/mL. It was applied with a flow rate of 10 μl/min onto a absorbance was measured at 280 nm in a Lambda Bio+ ultraviolet/visible (Perki- PepSwift trap column (200 μm × 5 mm, Thermo Fisher Scientific) in combination nElmer) spectrometer using a Quartz Suprasil R Ultra-Micro cuvette (Hellma). The with a monolithic ProSwift RP-4H analytical column (100 μm × 50 cm, Thermo protein concentration was determined based on the Lambert–Beer law using a Fisher Scientific), which was connected to a U3000 RSLCnano (Thermo Fisher −1 −1 theoretic molar extinction coefficient of 16,740 M cm for the FOR001 fibril Scientific) that was coupled to the mass spectrometer. The fibril protein was eluted protein according to the method of Gill and von Hippel . using a gradient of solvent B [86% (v/v) ACN, 0.1% (v/v) formic acid] and solvent A [0.1% (v/v) formic acid] with a flow rate of 1 μL/min. The gradient started with an increase of 5–55% solvent B over a period of 75 min, followed by an increase Fibril-formation kinetics measurements using ThT. Refolded and lyophilized from 55 to 95% over 15 min. The concentration of 95% solvent B stayed constant FOR001 fibril protein (glycoslylated or deglycosylated) was dissolved at 2 mg/mL for 3 minutes with a subsequent reduction from 95 to 5% solvent B over 9 min. concentration in water. Aggregation kinetics measurements were set up in PF 96- Fractions from the ProSwift RP-4H analytical column directly eluted into the well F-bottom black microplates (Greiner Bio-One International). Each well was ionization module and were further analyzed by MS. Samples were measured using filled with 100 μL of sample, containing 0.4 mg/mL glycosylated or deglycosylated an LTQ Orbitrap Elite system (Thermo Fisher Scientific). The mass spectrometer fibril protein, 20 μM ThT and 10 mM sodium acetate, 10 mM boric acid, 10 mM was equipped with a nanoelectrospray ion source and distal-coated SilicaTips sodium citrate, pH 4.0, and 150 mM NaCl. The plates were sealed with Rotilabo- (FS360-20-10-D, New Objective). The instrument was externally calibrated using sealing film (Carl Roth) and incubated at 37 °C in FLUOstar Omega (BMG Lab- standard compounds (LTQ Velos ESI Positive Ion Calibration Solution, Pierce, tech) for 72 h. During incubation, the plates were agitated by orbital shaking at Thermo Scientific) and operated using the following parameters: spray voltage, 300 rpm, which was paused during measurement. The fluorescence emission 1.5 kV; capillary temperature, 250 °C; S-lens radio-frequency level, 68.9%. The intensity at 490 nm was recorded every 30 min upon excitation at 450 nm. software XCalibur 2.2 SP1.48 (Thermo Fisher Scientific) was used for data- dependent MS/MS analyses. Full scans ranging from mass-to-charge ratio (m/z) Proteinase K digestion of amyloid fibrils. Aliquots of solutions containing ex vivo 370–1700 were acquired in the Orbitrap at a resolution of 30,000 (at m/z 400) with FOR001 amyloid fibrils or in vitro formed fibrils from glycosylated or deglycosylated automatic gain control enabled and set to 10 ions and a maximum fill time of FOR001 fibril protein (from the ThT kinetic experiment) were mixed with water and 500 ms. The raw data were deconvoluted by the MASH Explorer using default a 10 × buffer stock [200 mM Tris, pH 8.0, 1.4 M NaCl, 20 mM CaCl , and 1% (w/v) settings and the Quick Deconvolution feature. All calculated monoisotopic masses NaN ] to reach a total volume of 60 μL containing 0.2 mg/mL protein in 1 × buffer. A with a score equal to or above 94% resulting from initial m/z peaks with 5 charges first aliquot (10 μL) was withdrawn from this solution and retained for gel electro- or more were considered as correct and are shown in Supplementary Fig. 5b, c. The phoresis as the control sample without protease. The remaining 50 μL of the protein deconvoluted mass peaks were further assigned to protein species by using the solution were mixed with 1 μL of a 2 mg/mL proteinase K solution (Thermo Fisher software mMass considering a tolerance of 0.1 Da and a peak charge of 0. Scientific). Immediately afterward, a second aliquot (10 μL) was removed (0-min Sequence modifications were set as follows: pyroglutamylation at Gln1 was set as sample). The remaining solution was incubated at 37 °C in a heating block and variable, whereas the disulfide bond between Cys22 and Cys89 was set as fixed. further aliquots (10 μL) were withdrawn after 1 min, 2 min, and 5 min. As soon as an aliquot was withdrawn, it was mixed with 1 μL of 200 mM phenylmethylsulfonyl fluoride (PMSF) (Carl Roth) in methanol, incubated for 1 min at room temperature, Refolding of the FOR001 fibril protein. Solid guanidine hydrochloride was added and flash-frozen in liquid nitrogen. After the experiment, all aliquots were brought to to a sample of ex vivo FOR001 fibrils to reach a final concentration of 6 M followed room temperature and analyzed by denaturing protein-gel electrophoresis. The by an overnight incubation at room temperature to disaggregate the fibrils. The resulting protein bands (72 × 150 pixels) were densitometrically quantified using the protein was refolded by dialysis (molecular weight cutoff 3.5 kDa, Spectra/Por 6 program Fiji v1.52 . The intensity of the fibril protein band without proteinase K Dialysis Membrane Pre-wetted RC Tubing, Spectrum Labs) against 20 mM Tris was set to 100%, and an equally sized area on the gel without protein at 0%. buffer, pH 8.0, for 24 h at 4 °C. The protein was purified by anion-exchange chromatography with Q-SepharoseFF medium (10 mL, Cytiva) in an XK 16/20 column (Cytiva) with a slope gradient from 0% to 100% elution buffer [20 mM Tris Denaturing protein gel electrophoresis. Samples from the deglycosylation buffer, 1 M NaCl, pH 8.0] over 20 column volumes (CVs). The fibril protein- experiment and proteolytic stability measurement (10-μL volume for deglycosy- containing fractions, as identified by protein-gel electrophoresis, were purified lation, 11-μL volume for proteolytic stability, including 1-μL PMSF) were mixed further with a Resource 15 RPC column (3 mL, Cytiva) that was equilibrated in with 2 μL of 10 × NuPAGE reducing agent (Thermo Fisher Scientific), 5 μLof solvent A [0.1% (v/v) trifluoroacetic acid (TFA) in water]. The protein was eluted 4 × NuPAGE LDS sample buffer (Thermo Fisher Scientific), and water to generate through a slope gradient from 0 to 58% solvent B [86% (v/v) acetonitrile (ACN), a sample with a total volume of 20 μL. The solution was heated at 95 °C for 10 min 0.1% (v/v) TFA] over 20 CVs, followed by second gradient from 58 to 100% solvent and applied onto a 4–12% NuPAGE Bis-Tris gel (Thermo Fisher Scientific), B over 4 CVs to remove other bound proteins. Fractions were collected and the operated in NuPAGE MES SDS running buffer (Thermo Fisher Scientific). Blue- fractions containing the FOR001 fibril protein were identified by protein gel Easy Prestained (Genetics) was used as a marker. The gel was stained in a solution electrophoresis, pooled, and lyophilized. containing 30% (v/v) ethanol, 10% (v/v) acetic acid, and 0.25% (w/v) Coomassie brilliant blue and destained with a solution containing 20% (v/v) ethanol and 10% (v/v) acetic acid. Deglycosylation of the refolded FOR001 fibril protein. For the experiment shown in Supplementary Fig. 5a, the lyophilized, refolded FOR001 fibril protein was dissolved in water at approximately 2 mg/mL concentration. The exact protein Analysis of GL segments and mutations. The FOR001 amino acid sequence was concentration was determined by the intrinsic protein absorbance at 280 nm. This used to search the IgBLAST database (https://www.ncbi.nlm.nih.gov/igblast/), solution was mixed with water and SDS-containing 10 × glycoprotein-denaturing which returned the GL segment IGLV1-51*02. The protein sequence of the GL buffer [5% (w/v) SDS, 400 mM dithiolthreitol] (New England Biolabs) to generate a segment was retrieved from the VBase2 database (http://www.vbase2.org/). The final FOR001 fibril protein solution with the volume V containing a protein FOR001 J segment was compared with the five functional IGLJ GL segments in the Prot concentration of 1 mg/mL and 1 x Glycoprotein Denaturing Buffer (New England GenBank database (https://www.ncbi.nlm.nih.gov/genbank/). It matched both Biolabs). The protein was heated for 10 min at 100 °C to partially denature the IGLJ2 (Gene ID: 28832) and IGLJ3 (GeneID: 28831) and no unique source could be protein before it was cooled to room temperature. For N-deglycosylation, the determined. Nor could we identify the GL precursor of the C segment as most of heated sample was mixed with 1/5 V of a 10% (v/v) solution of the detergent the C domain is missing in the FOR001 fibril protein. The residues Leu97–Ala98 Prot L Nonident P40 in water (New England Biolabs), 1/5 V 10 × Glycobuffer 2 constitute the variable V/J junctional region. CDRs were determined with abYsis Prot NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 (http://www.abysis.org/abysis/) based on the Kabat definition . In this paper, all 9. Abraham, R. S. et al. Immunoglobulin light chain variable (V) region genes mutations are represented in the direction GL to fibril protein. influence clinical presentation and outcome in light chain–associated amyloidosis (AL). Blood 101, 3801–3807 (2003). 10. Kourelis, T. V. et al. Clarifying immunoglobulin gene usage in systemic and Computation of the aggregation score. To compute the aggregation score, we localized immunoglobulin light-chain amyloidosis by mass spectrometry. 69 70 71 72 used TANGO version 2.1 , WALTZ , FoldAmyloid , Aggrescan , and 73 Blood 129, 299–306 (2017). PASTA2.0 . These programs calculate a residue-specific aggregation potential. 11. Perfetti, V. et al. The repertoire of lambda light chains causing predominant The following settings were chosen for each program: TANGO: temperature: amyloid heart involvement and identification of a preferentially involved 309.15 K, ionic strength: 0.02 M, and concentration: 1 M and pH 7.0; WALTZ: the germline gene, IGLV1-44. Blood 119, 144–150 (2012). threshold was set to high sensitivity and the pH to 7.0; FoldAmyloid: scale: triple 12. Hurle, M. R., Helms, L. R., Li, L. I. N., Chan, W. & Wetzel, R. A role for hybrid, averaging frame: 5; Aggrescan: default settings; PASTA: 90% specificity, top destabilizing amino acid replacements in light-chain amyloidosis. Proc. Natl. pairing energies: 22. Residues with a high aggregation potential are defined as Acad. Sci. USA 91, 5446–5450 (1994). follows: TANGO: β-sheet aggregation values above 0.0; WALTZ: total sequence 13. Wall, J. et al. Thermodynamic instability of human lambda 6 light chains: score above 0.0; Foldamyloid: five successive amino acids with a triple-hybrid threshold above 0.062; Aggrescan: aggregation-propensity values above −0.02; Correlation with fibrillogenicity Biochemistry. Biochem. 38, 14101–14108 (1999). PASTA: PASTA energy units below −2.8. An aggregation score of 0 means that none of the programs identifies a high aggregation score for a given residue. An 14. Oberti, L. et al. Concurrent structural and biophysical traits link with aggregation score of 5 means that all five programs predict a high aggregation score immunoglobulin light chains amyloid propensity. Sci. Rep. 7,1–11 (2017). for that residue. 15. Kazman, P. et al. Fatal amyloid formation in a patient’s antibody light chain is caused by a single point mutation. eLife 9, e52300 (2020). 16. Rottenaicher, G. J. et al. Molecular mechanism of amyloidogenic mutations in Protein structure representation. The images of the density map and protein 74 hypervariable regions of antibody light chains. J. Biol. Chem. 296, 100334 model were created with the software UCSF Chimera v1.14 . Hydrogen bonds (2021). were defined according to the criteria of both UCSF Chimera v.1.14 and the 61 17. Blancas-Mejía, L. M. et al. Kinetic control in protein folding for light chain software Coot v0.8.9 . The β-sheets were defined as a minimum of two residues amyloidosis and the differential effects of somatic mutations. J. Mol. Biol. 426, having φ/ψ angles in the β-sheet region of the Ramachandran plot and at least one- 347–361 (2014). backbone hydrogen bond. 18. Piehl, D. W., Blancas-Mejía, L. M., Ramirez-Alvarado, M. & Rienstra, C. M. Solid-state NMR chemical shift assignments for AL-09 V L immunoglobulin Sample statistics. In this paper errors report the standard deviation. In Fig. 5e, a light chain fibrils. Biomol. NMR Assign. 11,45–50 (2017). one-tailed Welch t-test was used. 19. Pradhan, T. et al. Seeded fibrils of the germline variant of human λ-III immunoglobulin light chain FOR005 have a similar core as patient fibrils with Reporting summary. Further information on research design is available in the Nature reduced stability. J. Biol. Chem. 295, 18474–18484 (2020). Research Reporting Summary linked to this article. 20. Annamalai, K. et al. Common fibril structures imply systemically conserved protein misfolding pathways in vivo. Angew. Chem. 129, 7618–7622 (2017). 21. Radamaker, L. et al. Cryo-EM structure of a light chain-derived amyloid Data availability fibril from a patient with systemic AL amyloidosis. Nat. Commun. 10.1,1–8 The datasets used during the current study are available from public repositories and/or (2019). from the corresponding author on reasonable request. The cryo-EM map of the FOR001 22. Swuec, P. et al. Cryo-EM structure of cardiac amyloid fibrils from an fibrils was deposited in the Electron Microscopy Data Bank (https://www.ebi.ac.uk/pdbe/ immunoglobulin light chain AL amyloidosis patient. Nat. Commun. 10.1,1–9 emdb/) with the accession code EMD-12570. The coordinates of the corresponding (2019). atomic model were deposited in the PDB (https://www.rcsb.org/) under the accession 23. Radamaker, L. et al. Cryo-EM reveals structural breaks in a patient-derived code 7NSL. The cryo-EM data of the FOR001 fibrils were deposited on EMPIAR (https:// amyloid fibril from systemic AL amyloidosis. Nat. Commun. 12,1–10 (2021). www.ebi.ac.uk/pdbe/emdb/empiar/) with the accession code EMPIAR-10730. The 24. Yazaki, M., Liepnieks, J. J., Callaghan, J., Connolly, C. E. & Benson, M. D. following published PDB structures were used in the paper: 4ODH, 6IC3, 6Z1O, 6HUD, Chemical characterization of a lambda I amyloid protein isolated from 5JZ7, 6QB6, 6Q0E, 7JVA, and 5MUD. The accession codes for the IGLV1-51*02 segment formalin-fixed and paraffin-embedded tissue sections. Amyloid 11,50–55 from the IMGT database are M30446 and from the VBase2 humIGLV015. The IGLJ2 (2004). and IGLJ3 gene segments were taken from GenBank with the Gene IDs 28832 and 28831, 25. Lu, Y., Jiang, Y., Prokaeva, T., Connors, L. H. & Costello, C. E. Oxidative post- respectively. Source data are provided with this paper for the following figures: Fig. 1c, translational modifications of an amyloidogenic immunoglobulin light chain Fig. 5c, d, e, Supplementary Fig. 1a, Supplementary Fig. 2, Supplementary Fig. 5d, and protein. Int. J. Mass Spectrom. 416,71–79 (2017). Supplementary Fig. 6. Source data are provided with this paper. 26. Kumar, S. et al. Assay to rapidly screen for immunoglobulin light chain glycosylation: a potential path to earlier AL diagnosis for a subset of patients. Leukemia 33, 254–257 (2019). Received: 18 June 2021; Accepted: 1 October 2021; 27. Mellors, P. W. et al. MASS-FIX for the detection of monoclonal proteins and light chain N-glycosylation in routine clinical practice: a cross-sectional study of 6315 patients. Blood Cancer J. 11,1–9 (2021). 28. Dispenzieri, A. et al. N-glycosylation of monoclonal light chains on routine MASS-FIX testing is a risk factor for MGUS progression. Leukemia 34, 2749–2753 (2020). References 29. Sagis, L. M., Veerman, C. & van der Linden, E. Mesoscopic properties of 1. Blancas-Mejia, L. M. et al. Immunoglobulin light chain amyloid aggregation. semiflexible amyloid fibrils. Langmuir 20, 924–927 (2004). Chem. Commun. 54.76, 10664–10674 (2018). 30. Adamcik, J. & Mezzenga, R. Study of amyloid fibrils via atomic force 2. Herrera, G. A., Teng, J., Turbat-Herrera, E. A., Zeng, C. & del Pozo-Yauner, L. microscopy. Curr. Opin. Colloid Interface Sci. 17, 369–376 (2012). Understanding Mesangial Pathobiology in AL-amyloidosis and Monoclonal 31. Palladini, G. et al. Circulating amyloidogenic free light chains and serum Immunoglobulin Light Chain Deposition Disease. Kidney Int. Rep. (2020). N-terminal natriuretic peptide type B decrease simultaneously in association 3. Wechalekar, A. D. et al. A European collaborative study of treatment with improvement of survival in AL. Blood 107, 3854–3858 (2006). outcomes in 346 patients with cardiac stage III AL amyloidosis. Blood 121, 32. Poshusta, T. L. et al. Mutations in specific structural regions of 3420–3427 (2013). immunoglobulin light chains are associated with free light chain levels in 4. Merlini, G. et al. Systemic immunoglobulin light chain amyloidosis. Nat. Rev. patients with AL amyloidosis. PloS ONE 4, e5169 (2009). Dis. Prim. 4,1–19 (2018). 33. Thal, D. R., Walter, J., Saido, T. C. & Fändrich, M. Neuropathology and 5. Janeway C. A. Jr., Travers, P., Walport, M. & Shlomchik, M. J. Immunobiology: biochemistry of Aβ and its aggregates in Alzheimer’s disease. Acta the Immune System in Health and Disease. 5th edn (Garland Science, 2001). Neuropathol. 129, 167–182 (2015). The generation of diversity in immunoglobulins. 34. Wulff, M. et al. Enhanced fibril fragmentation of N-terminally truncated and 6. Lefranc, M. P. IMGT®, the international ImMunoGeneTics database. Nucl. pyroglutamyl-modified Aβ peptides. Angew. Chem.Int.Ed. 55,5081–5084 (2016). Acids Res. 29, 207–209 (2001). 35. Stevens, F. J. Four structural risk factors identify most fibril-forming kappa 7. Comenzo, R. L., Zhang, Y., Martinez, C., Osman, K. & Herrera, G. A. The tropism light chains. Amyloid 7, 200–211 (2000). of organ involvement in primary systemic amyloidosis: contributions of Ig VL 36. Omtvedt, L. A. et al. Glycosylation of immunoglobulin light chains associated germ line gene use and clonal plasma cell burden. Blood 98,714–720 (2001). with amyloidosis. Amyloid 7, 227–244 (2000). 8. Perfetti, V. et al. Analysis of Vλ-Jλ expression in plasma cells from primary 37. Solá, R. J. & Griebenow, K. A. I. Effects of glycosylation on the stability of (AL) amyloidosis and normal bone marrow identifies 3r (λIII) as a new protein pharmaceuticals. J. Pharm. Sci. 98, 1223–1245 (2009). amyloid-associated germline gene segment. Blood 100, 948–953 (2002). 10 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE 38. Schwarz, F. & Aebi, M. Mechanisms and principles of N-linked protein 69. Fernandez-Escamilla, A. M., Rousseau, F., Schymkowitz, J. & Serrano, L. glycosylation. Curr. Opin. Struct. Biol. 21, 576–582 (2011). Prediction of sequence-dependent and mutational effects on the aggregation 39. Lee, H. S., Qi, Y. & Im, W. Effects of N-glycosylation on protein conformation of peptides and proteins. Nat. Biotechnol. 22, 1302–1306 (2004). and dynamics: Protein Data Bank analysis and molecular dynamics simulation 70. Maurer-Stroh, S. et al. Exploring the sequence determinants of amyloid study. Sci. Rep. 5,1–7 (2015). structure using position-specific scoring matrices. Nat. Methods 7, 237–242 40. Bellotti, V., Mangione, P. & Merlini, G. Immunoglobulin light chain (2010). amyloidosis—the archetype of structural and pathogenic variability. J. Struct. 71. Garbuzynskiy, S. O., Lobanov, M. Y. & Galzitskaya, O. V. FoldAmyloid: a Biol. 130, 280–289 (2000). method of prediction of amyloidogenic regions from protein sequence. 41. Pepys, M. B. et al. Human serum amyloid P component is an invariant Bioinformatics 26, 326–332 (2010). constituent of amyloid deposits and has a uniquely homogeneous 72. Conchillo-Solé, O. et al. AGGRESCAN: a server for the prediction and glycostructure. Proc. Natl. Acad. Sci. USA 91, 5602–5606 (1994). evaluation of “hot spots” of aggregation in polypeptides. BMC Bioinform. 8,65 42. Liberta, F. et al. Morphological and primary structural consistency of fibrils (2007). from different AA patients (common variant). Amyloid 26, 164–170 (2019). 73. Walsh, I., Seno, F., Tosatto, S. C. & Trovato, A. PASTA 2.0: an improved 43. Schmidt, M. et al. Cryo-EM structure of a transthyretin-derived amyloid fibril server for protein aggregation prediction. Nucl. Ac. Res. 42, W301–W307 from a patient with hereditary ATTR amyloidosis. Nat. Commun. 10,1–9 (2019). (2014). 44. Scheres, S. H., Zhang, W., Falcon, B. & Goedert, M. Cryo-EM structures of tau 74. Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory filaments. Curr. Opin. Struct. Biol. 64,17–25 (2020). research and analysis. J. Comput. Chem. 25, 1605–1612 (2004). 45. Bergström, J. et al. Amyloid deposits in transthyretin derived amyloidosis: cleaved transthyretin is associated with distinct amyloid morphology. J. Acknowledgements Pathol. 206, 224–232 (2005). We thank the Deutsche Forschungsgemeinschaft for funding (projects FA 456/23 and FA 46. Westermark, G. T., Fändrich, M. & Westermark, P. AA amyloidosis: 456/27 to M.F., HA 7138/3 to C.H., HE 8472/1-1 to U.H., HU 2400/1-1 to S.H., SCHO pathogenesis and targeted therapy. Annu. Rev. Pathol. 10, 321–344 (2015). 1364/2-1 to S. O. S., and SCHM 997/30-1 to V.S and a Start-up Grant of the Research 47. Kollmer, M. et al. Cryo-EM structure and polymorphism of Aβ amyloid fibrils Unit FOR 2969 to C.R.). All cryo-EM data were collected at the European Molecular purified from Alzheimer’s brain tissue. Nat. Commun. 10,1–8 (2019). Biology Laboratory, Heidelberg (Germany), funded by iNEXT (Horizon 2020, European 48. Schweighauser, M. et al. Structures of α-synuclein filaments from multiple Union). We are grateful to Karl-Heinz Gührs for critically reading the paper. system atrophy. Nature 585, 464–469 (2020). 49. Bansal, A. et al. AA amyloid fibrils from diseased tissue are structurally different from in vitro formed SAA fibrils. Nat. Commun. 12,1–9 (2021). Author contributions 50. Fändrich, M. & Schmidt, M. Methods to study the structure of misfolded protein L.R., S.K., G.A., J.B., S.W., and C.R. carried out experiments. L.R, S.K. G.A., J.B., M.N., states in systemic amyloidosis. Biochemical Soc. Trans. 49, 977–985 (2021). S.S., N.B., C.H., V.S., S.H., P.W., S.W., C.R., M.S., and M.F. analyzed data. R.M., U.H., 51. Read, C., Walther, P. & von Einem, J. Quantitative Electron Microscopy to and S.O.S. contributed tools, reagents, or materials. M.F. designed research. L.R. and M.F. Study HCMV Morphogenesis, in: Human Cytomegaloviruses: Methods and wrote the paper with contributions from all other authors. Protocols, Methods in Molecular Biology (Ed. Yurochko, A. D.) 265–289 (Springer US, 2021). 52. Read, C., Schauflinger, M., Nikolaenko, D., Walther, P. & von Einem, J. Funding Open Access funding enabled and organized by Projekt DEAL. Regulation of human cytomegalovirus secondary envelopment by a C-terminal tetra-lysine motif in pUL71. J. Virol. 93, e02244-18 (2019). 53. Kremer, J. R., Mastronarde, D. N. & McIntosh, J. R. Computer visualization of Competing interests three-dimensional image data using IMOD. J. Struct. Biol. 116,71–76 (1996). The authors declare no competing interests. 54. Kollmer, M. et al. Electron tomography reveals the fibril structure and lipid interactions in amyloid deposits. Proc. Natl. Acad. Sci. USA 113, 5604–5609 (2016). Additional information 55. Feuz, L., Leermakers, F. A. M., Textor, M. & Borisov, O. Bending rigidity and Supplementary information The online version contains supplementary material induced persistence length of molecular bottle brushes: a self-consistent-field available at https://doi.org/10.1038/s41467-021-26553-9. theory. Macromolecules 38, 8891–8901 (2005). 56. Annamalai, K. et al. Polymorphism of amyloid fibrils in vivo. Angew. Chem. Correspondence and requests for materials should be addressed to Marcus Fändrich. Int. Ed. 55.15, 4822–4825 (2016). Peer review information Nature Communications thanks Gareth J Morgan, Luis del 57. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Pozo-Yauner and the other, anonymous, reviewer(s) for their contribution to the peer Nat. Methods 9, 676–682 (2012). review of this work. 58. Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017). 59. Zhang, K. Gctf: Real-time CTF determination and correction. J. Struct. Biol. Reprints and permission information is available at http://www.nature.com/reprints 193,1–12 (2016). 60. He, S. & Scheres, S. H. W. Helical reconstruction in RELION. J. Struct. Biol. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 198, 163–176 (2017). published maps and institutional affiliations. 61. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of coot. Acta Crystallogr. D. Biol. Crystallogr. 66, 486–501 (2010). 62. Afonine, P. V. et al. Real-space refinement in PHENIX for cryo-EM and Open Access This article is licensed under a Creative Commons crystallography. Acta Crystallogr. D. Struct. Biol. 74, 531–544 (2018). Attribution 4.0 International License, which permits use, sharing, 63. Afonine, P. V. et al. New tools for the analysis and validation of cryo-EM adaptation, distribution and reproduction in any medium or format, as long as you give maps and atomic models. Acta Crystallogr. D. Struct. Biol. 74, 814–840 (2018). appropriate credit to the original author(s) and the source, provide a link to the Creative 64. Williams, C. J. et al. MolProbity: more and better reference data for improved Commons license, and indicate if changes were made. The images or other third party all atom structure validation. Protein Sci. 27.1, 293–315 (2018). material in this article are included in the article’s Creative Commons license, unless 65. Wu, Z. et al. MASH explorer: a universal software environment for top-down indicated otherwise in a credit line to the material. If material is not included in the proteomics. J. Proteome Res. 19, 3867–3876 (2020). article’s Creative Commons license and your intended use is not permitted by statutory 66. Niedermeyer, T. H. & Strohalm, M. mMass as a software tool for the regulation or exceeds the permitted use, you will need to obtain permission directly from annotation of cyclic peptide tandem mass spectra. PloS ONE 7, e44913 (2012). the copyright holder. To view a copy of this license, visit http://creativecommons.org/ 67. Gill, S. C. & Von Hippel, P. H. Calculation of protein extinction coefficients licenses/by/4.0/. from amino acid sequence data. Anal. Biochem. 182, 319–326 (1989). 68. Kabat. E. A., Wu, T. T., Reid-Miller, M., Peny, H. & Gottesman, K. Sequences of Proteins of Immunological Interest. 4th edn (United States Department of © The Author(s) 2021 Health and Human Services, 1987) NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 11 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Loading next page...
 
/lp/springer-journals/role-of-mutations-and-post-translational-modifications-in-systemic-al-0gHCRVSIj7

References (83)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2041-1723
DOI
10.1038/s41467-021-26553-9
Publisher site
See Article on Publisher Site

Abstract

ARTICLE https://doi.org/10.1038/s41467-021-26553-9 OPEN Role of mutations and post-translational modifications in systemic AL amyloidosis studied by cryo-EM 1 1 1 1 2 Lynn Radamaker , Sara Karimi-Farsijani , Giada Andreotti , Julian Baur , Matthias Neumann , 3 3 4 1 5 2 Sarah Schreiner , Natalie Berghaus , Raoul Motika , Christian Haupt , Paul Walther , Volker Schmidt , 3 6 6 7 5,8 Stefanie Huhn , Ute Hegenbart , Stefan O. Schönland , Sebastian Wiese , Clarissa Read , 1 1 Matthias Schmidt & Marcus Fändrich Systemic AL amyloidosis is a rare disease that is caused by the misfolding of immunoglobulin light chains (LCs). Potential drivers of amyloid formation in this disease are post-translational modifications (PTMs) and the mutational changes that are inserted into the LCs by somatic hypermutation. Here we present the cryo electron microscopy (cryo-EM) structure of an ex vivo λ1-AL amyloid fibril whose deposits disrupt the ordered cardiomyocyte structure in the heart. The fibril protein contains six mutational changes compared to the germ line and three PTMs (disulfide bond, N-glycosylation and pyroglutamylation). Our data imply that the disulfide bond, glycosylation and mutational changes contribute to determining the fibril protein fold and help to generate a fibril morphology that is able to withstand proteolytic degradation inside the body. 1 2 3 Institute of Protein Biochemistry, Ulm University, 89081 Ulm, Germany. Institute of Stochastics, Ulm University, 89081 Ulm, Germany. Medical Department V, Section of Multiple Myeloma, Heidelberg University Hospital, 69120 Heidelberg, Germany. Department of Asia-Africa-Studies, Middle Eastern History and Culture, University of Hamburg, 20148 Hamburg, Germany. Central Facility for Electron Microscopy, Ulm University, 89081 6 7 Ulm, Germany. Medical Department V, Amyloidosis Center, Heidelberg University Hospital, 69120 Heidelberg, Germany. Core Unit Mass Spectrometry and Proteomics, Medical Faculty, Ulm University, 89081 Ulm, Germany. Institute of Virology, Ulm University Medical Center, 89081 Ulm, Germany. email: marcus.faendrich@uni-ulm.de NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ystemic AL amyloidosis is defined by the formation of fibrils that show a width of ~9 nm. Quantification of the fibril 1,2 amyloid fibrils by immunoglobulin light chains (LCs) . end-to-end distances and contour lengths allowed us to deter- SThese fibrils can deposit at multiple sites in the body where mine the fibril persistence length at 0.74 ± 0.08 μm and its −27 −28 2 they can lead to severe impairment of vital organ functions. bending rigidity at 3.1 × 10 ±3× 10 Nm (Fig. 1c). The Untreated patients with a prominent heart involvement show a persistence length corresponds to values reported in the 29,30 high risk of death with a median survival of only seven months literature for amyloid fibrils, which vary between 16 nm and 3,4 after their initial diagnosis . Due to their natural function as part 18.5 μm, indicating that FOR001 amyloid fibrils and their of B-cell receptors and antibodies, LCs are hypervariable proteins. deposits are structurally rigid. The fibrils in these deposits interact This variability arises from the genetic recombination of variable with the surfaces of adjacent cardiomyocytes, and these interac- (V), joining (J), and constant (C) gene segments that are encoded tions occur mainly via the fibril tips (Fig. 1d), and only rarely via in the germ line (GL), somatic hypermutation, and the junctional the fibril lateral surfaces. In some cases, we find deformations in diversity at the V/J interface in the course of V/J recombination . the plasma membrane associated with the focal contact points of Altogether, there are 63–71 functional V (34–38 V and 29–33 the fibrils (Fig. 1d). The fibrils impede the contractile function of V ), 9–10 J (5 J and 4–5J ), and 5–6 C (1 C and 4–5C ) the heart, suggesting that clearance of the amyloid may help to λ κ λ κ λ segments encoded within the GL of which IGLV6-57, IGLV3-01, restore cardiac function. While there is evidence that the patient’s IGLV2-14, and IGKV1-33 were found to be overrepresented in health condition is also defined by circulating amyloid precursors 7–10 31 systemic AL amyloidosis . These findings and the fact that in the serum , our observations underpin the view that the usage of IGLV1-44 is associated with cardiac involvement imply amyloid deposits are damaging to the patient. that the LC primary structure is a key determinant for the development of amyloidosis. Further support for this view comes from observations that Cryo-EM structure of the FOR001 AL amyloid fibril.To the mutational changes that are inserted into the LCs during investigate the structure of the extracted FOR001 AL fibrils, we somatic hypermutation affect the kinetics of amyloid fibril for- imaged them using cryo-EM (Fig. 2a). The images show that mation in vitro . The effect of these mutations has frequently approximately 75% of the fibrils seen in the micrographs belong been attributed to alterations in the biophysical properties to one dominant morphology. This fibril morphology is defined of the natively folded, globular LCs, such as to a decreased by a width of ~9 nm, which agrees with the width measured by 12,13 thermodynamic stability , to increased conformational electron tomography (see above), and a crossover distance of 14–16 dynamics , or to local structural changes in a key region of ~55 nm, as measured from the recorded images. 3D reconstruc- 14–16 the protein . In other cases, it was suggested that the muta- tion of the fibril images yielded a 3D map of the dominant tions may affect the formation of specific folding or misfolding morphology with a resolution of 3.1 Å (Supplementary Fig. 1a, 14,17 intermediates or the stability or structure of the resulting Supplementary Table 1), while the remaining minor morpholo- 18,19 amyloid fibril . Post-translational modifications (PTMs) may gies could not be reconstructed. The rise and twist of the further modulate the effects of the inserted mutations. Several reconstructed fibril are 4.76 Å and −1.46°, the pitch is 117 nm studies have shown that AL fibril proteins can contain PTMs like (Supplementary Table 1). The fibril consists of a single proto- 20–23 disulfide bonds , N-terminal pyroglutamate (Pyro-Glu) filament (C1 symmetry, Fig. 2b, c). The fitted molecular model 24,25 26,27 modifications , or glycosylations . Glycosylation was found (Fig. 2d) has a model resolution of 3.1 Å, (Supplementary to be overrepresented in AL patients, indicating that this PTM Table 1). Projections of its density onto the y–z plane correspond 26,28 contributes to the pathogenesis in AL amyloidosis . Yet, the well to the two-dimensional (2D) class averages (Supplementary mechanism by which glycosylation affects fibril formation or AL Fig. 1b). The handedness of the fibrils in this sample was deter- pathogenesis has so far remained unclear. mined by platinum side-shadowing, which showed a left-handed To shed light on the structural effects of PTMs and mutational twist (Supplementary Fig. 2). The ordered core of the fibril changes on the fibril state, we have determined the structure of an consists of two segments that extend from Ser9 to Thr52 and AL amyloid fibril with cryoelectron microscopy (cryo-EM), which Ser68 to Thr108 (Fig. 2d). For residues Gln1–Pro8, Asp53–Lys67, is partially pyroglutamylated, N-glycosylated and modified by an and Val109–Ser118, which are present in the fibril as shown by intramolecular disulfide bond. The observed fibril structure is mass spectrometry (MS) (see below), no well-defined density different from previously described, nonglycosylated AL amyloid could be discerned in our map, suggesting that they are struc- 21–23 fibrils , consistent with the patient-specific nature of this turally disordered. The fibril protein contains an intramolecular disease. The mutational changes are clustered into two topolo- disulfide that connects residues Cys22 and Cys89 (Fig. 2d). gical regions of natively folded variable LC domains, but dis- persed throughout the fold of the fibril protein with no obvious structural preference. Chemical interactions stabilizing the fibril protein fold.Despite being dominated by β-sheet conformation, the fibril protein structure is profoundly different from the structure of a natively folded LC Results (Fig. 3a) and encompasses eleven parallel cross-β-sheets (β1–β11) FOR001 amyloid fibrils are structurally rigid and disrupt the that are formed by residues Val10–Ala12, Pro14–Ser21, ordered architecture of the heart muscle. The presently analyzed Asn31–Val34, Tyr37–Gln39, Thr43–Ala44, Pro45–Glu51, fibrils were extracted from the heart tissue of patient FOR001, Thr70–Leu74, Ile76–Gly78, Tyr88–Cys89, Thr91–Glu93, and who suffered from advanced cardiac AL amyloidosis and Thr105–Leu107 (Fig. 3a, b). The β-strands in these sheets interact in underwent a heart transplantation at the age of 51 years. Analysis the direction of the fibril z axis through backbone hydrogen bonds as of sections of cardiac tissue with scanning electron microscopy well as through side-chain interactions, including the stacking of (SEM) demonstrates that the fibrils form large-sized, extracellular aromatic and polar residues (Supplementary Fig. 3a, b). The fibril amyloid deposits that infiltrate and disrupt the ordered structure protein fold is defined by buried electrostatic interactions, such as of the cardiomyocytes (Fig. 1a, b). Using scanning transmission between Glu84 and Lys46 (Supplementary Fig. 3c), and by several electron microscopy (STEM), we obtained tomograms of the buried patches of hydrophobic residues, such as the one formed by three-dimensional (3D) structure of the cardiac amyloid deposits residues Val10, Ala12, Val34, and Trp36 (Fig. 3c). The fibril qua- (Fig. 1b). The deposits are composed of haphazardly arranged ternary structure is stabilized by an interlocking of the protomers in 2 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE ab 0.15 measured data points fitted curve 0.1 0.05 0 0.1 0.2 0.3 0.4 contour length [μm] Fig. 1 Tissue deposits of FOR001 amyloid fibrils. a SEM overview of FOR001 heart tissue. The fibril deposits between cardiomyocytes are marked with blue asterisks. Scale bar: 1 μm. b STEM tomogram. Top: virtual section. The fibril deposit is marked with a blue asterisk. Bottom: rendered tomogram of the fibril deposit. Blue: amyloid fibrils. Red: membranes. Scale bar: 100 nm. c Analysis of the persistence length based on a fit of the plot showing the squared end-to-end distance versus the contour length using Eq. (1). The blue symbols show the measured data points (n = 195) and the red line the fit. d Region of the tomogram, showing the interactions of the fibrils (blue) with the cardiomyocyte membrane (red). Scale bar 100 nm. the direction of the z-axis that is caused by a 14-Å height change of immunoglobulin LCs and their GL segments (see “Methods” for the fibril protein backbone (Supplementary Fig. 3d). details) revealed that the FOR001 fibril protein stems from the GL segment IGLV1-51*02. The GL precursors of the J and C seg- ments could not be identified unambiguously. In case of the C Primary structure of the FOR001 fibril protein. The FOR001 segment, the number of residues (8, Gly111–Ser118) was too fibril protein sequence was determined with electrospray- small to allow any GL assignment. In case of the J segment, ionization MS (Supplementary Fig. 4). DNA sequencing of IGLJ2*01, IGLJ3*01 (which are identical) and IGLJ3*02 are all bone-marrow-derived cDNA was also attempted but failed to possible precursors (IGLJ3*02 differs from IGLJ2*01 and produce a sequence that matched the sequence obtained by direct IGLJ3*01 only in the first residue, which is mutated in the sequencing of the fibril protein. Three major fibril protein species FOR001 protein sequence), also precluding a unique GL assign- were revealed by MS (Supplementary Fig. 5a). Two of these could ment. The FOR001 sequence shows six mutational changes be assigned to the LC fragments Pyro–Glu1 to Ser118 and Ser2 to compared with the GL segments IGLV1-51*02, IGLJ2*01, Ser118 (Supplementary Fig. 5b), both containing a disulfide bond. IGLJ3*01, and IGLJ3*02: Lys17Asn, Asn52Thr, Asn53Asp, The third species corresponds to the fragment containing residues Gly82Ala, and Asp93Glu in the V segment and Xaa99Gly in the J Val3–Ser118 and a disulfide bond, although alternative mass segment (Fig. 4a, b, indigo). The mutation Lys17Asn inserts the assignments are also possible for the recorded MS peak (Sup- glycosylation site. In addition, residues Leu97–Ala98 form the plementary Fig. 5b). These analyses reveal two PTMs: a disulfide variable junction between the V and J segments that arises from bond and a pyroglutamylation (Supplementary Fig. 5c) that genetic V/J recombination (Fig. 4a, b, green). affects only a fraction of the fibril proteins. The third PTM of the fibril protein is an ~2-kDa N-glycosylation, which is demon- strated by an electrophoretic band shift of the refolded FOR001 The location of the mutations in known fibril structures and in fibril protein upon addition of N-glycosidase but not with natively folded V domains. To identify the possible role of O-glycosidase (Supplementary Fig. 5d). The carbohydrate can be mutations in LC aggregation, we analyzed their position within seen as extra density within our 3D map that could not be the known fibril structures and natively folded LCs. We find that assigned to the polypeptide chain (Fig. 2b, d, red star). It pro- the majority of the mutations occur at solvent-exposed positions trudes from residue Asn17, which forms part of the only cano- in the fibril, specifically in the FOR001 fibril. Five FOR001 nical N-glycosylation site (Asn–Xaa–Thr/Ser) in the fibril protein mutations (Lys17Asn, Asn52Thr, Gly82Ala, Asp93Glu, and sequence. Xaa99Gly) are part of the ordered fibril core, the sixth mutation (Asn53Asp) affects a structurally disordered region (Fig. 4a, b). Identification of the GL segments and mutational changes. One residue (position 53 in the FOR001 fibril protein) is mutated Comparisons of the FOR001 LC with known sequences of in all known AL fibril structures (Fig. 4a, b) and was identified NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 3 squared end-to-end distance [μm²] ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 Fig. 2 Cryo-EM structure of the FOR001 AL amyloid fibril. a Cryo-EM image of FOR001 amyloid fibrils. Scale bar is 100 nm. The dataset consists of 3033 micrographs. b Cross section of the map obtained by summing five central slices. c Side view of the map (left) and molecular model (right), showing the left-handed fibril twist. d Cross section of the map with the molecular model overlaid. The color coding of the model is the same in panels (c) and (d), that is, light blue refers to the N-terminal segment of the ordered fibril protein (residues Ser9–Thr52), while the deep-red segment refers to the C-terminal segment (residues Ser68–Thr108). The two segments are cross-linked through a disulfide between Cys22 and Cys89. The red star in (b) and (d) indicates the glycosylation site at Asn17. Fig. 3 Location of the secondary structural elements and mutational sites in the fibril structure. a Amino acid sequence of the FOR001 fibril protein and secondary structural elements of the FOR001 fibril protein (PDB 7NSL) and of a crystal structure of a natively folded LC (PDB 4ODH 10.2210/pdb4ODH/ pdb) containing an IGLV1-51*02 segment. Arrows indicate β-strands in the structure, rainbow-colored from N (blue) to C terminus (red). Dotted lines represent disordered segments. Red star: location of the glycosylation. b Stack of seven protein layers of the fibril, showing the β-strands β1–β11 with the same coloring as in (a). c Schematic representation of the fibril protein fold. Red star: location of the glycosylation. 4 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE Fig. 4 Location of the mutations in known AL amyloid fibrils and natively folded V domains. a Location of the mutations in known AL amyloid fibril structures. The fibrils are derived from the GL segments IGLV1-51*02 λ1 (FOR001, this study, PDB 7NSL) IGLV1-44*01 λ1 (FOR006, PDB 6IC3 10.2210/ 21 23 22 pdb6IC3/pdb) , IGLV3-19*01 λ3 (FOR005, PDB 6Z1O 10.2210/pdb6Z1O/pdb) , and IGLV6-57*02 λ6 (AL55, PDB 6HUD 10.2210/pdb6HUD/pdb) . Disordered parts of the fibril proteins are indicated by dotted lines. Black: CDRs; yellow: Cys. Indigo: mutations in the CDRs and one residue before or after a CDR; magenta: mutations in framework regions; green: residues in the junctional region at the V/J interface. Red star: location of the glycosylation. b Sequence alignment of the four fibril proteins. CDRs are marked with gray boxes. Color coding as in (a). c Location of mutations in the corresponding, natively folded LC V domains that are based on the GL segments IGLV1-51*01 (PDB 5JZ7 10.2210/pdb5JZ7/pdb), IGLV1-44*01 (PDB 6QB6 10.2210/ pdb6QB6/pdb) IGLV3-19*01 (PDB 6Q0E 10.2210/pdb6Q0E/pdb) and IGLV6-57*02 (PDB 7JVA 10.2210/pdb7JVA/pdb) CDRs are marked in black. Color coding as in (a). previously as a mutational hotspot in systemic AL amyloidosis . fibril, which does not contain an internal disordered segment, the It is located in the central disordered segments that are present in PDRFSGS motif is not preceded by a segment with an the FOR001, FOR005, and AL55 fibril structures (Fig. 4a), indi- aggregation score of 5 (Supplementary Fig. 6a, b). These data cating that this residue does not substantially affect the fibril suggest that mutations and aggregation-prone segments help to stability. About 79 ± 21% of the mutations of the four known AL define the specific fold of the observed fibril morphologies. amyloid fibrils reside within an ordered structural segment of the The mutations may additionally affect the native state, as they are V domain (Supplementary Table 2). This value is identical, found to be clustered into two regions of the globular V domains: L L within error, to the percentage of residues forming the ordered One region is formed by the complementarity-determining region fibril parts (77 ± 6%), suggesting that the mutations are not pre- (CDR) mutations around the lower-right rim of the native fold when ferentially located in the fibril core. However, mutations within it is oriented as in Fig. 4c (blue spheres). This part of the structure is the ordered part of the fibril protein are generally well- involved in forming the antigen-binding site and includes the shared accommodated in the structure and can add specific interac- mutated residue (position 53 in FOR001). The second region is tions. In the FOR006 fibril, they enable the interactions between formed by the framework mutations in the upper part of the V Arg25 (mutated from Ser) and Glu84 [21]; in the FOR001 fibril, domain (Fig. 4c, magenta spheres). Both clusters correlate with between Ala82 (mutated from Gly) and Leu48. previously described mutational hotspots of AL fibril proteins . In a next step, we analyzed the position of the aggregation- Taken together with the observations above, we conclude that the prone segments in the fibril structures. These segments were mutational changes in AL patients preferentially affect two structural 21,23 identified here based on their theoretic aggregation score . sites in the native protein state. In addition, they contribute The aggregation-prone segments (as defined by an aggregation interactions to support the formation of a specific fibril morphology. score of 5) occur within the stable core of the FOR001 fibril and other known AL amyloid fibrils (Supplementary Fig. 6a, b). By Effect of the N-glycosylation on fibril formation. The glycosy- contrast, the structurally disordered regions of these fibrils lation site is exposed on the fibril surface (Fig. 2d) and on the typically show an aggregation score of 0 (Supplementary Fig. 6b). surface of a natively folded LC (Fig. 5a, b). In the native LC, it is The fibrils with an internal disordered segment (FOR001, far away from the interface to the heavy chain (Fig. 5a) or to the FOR005, and AL55) share a conserved sequence motif second LC in a LC dimer (Fig. 5b). These observations indicate (PDRFSGS) with low aggregation propensity that is that the glycosylation does not strongly interfere with the ability N-terminally preceded by a highly aggregation-prone segment of the FOR001 LC to assemble into antibodies or LC dimers. To (aggregation score 5, Supplementary Fig. 6a, b). In the FOR006 investigate the influence of the glycosylation on fibril formation NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 a b c V 80 V V H L L C 0 10 20 30 40 50 60 70 time (h) In vitro In vitro Ex vivo 100 glycosylated deglycosylated M M M - - - - - - P Proteinase K + + + + + + + + + + + + 80 80 62 62 50 50 31 31 20 4 5 0 1 2 3 6.5 6.5 6.5 time (min) Fig. 5 Effect of glycosylation on the formation of fibrils from FOR001 fibril protein. a Ribbon representations of a fragment antigen binding that contains a LC with an IGLV1-51*02 GL segment (PDB 4ODH 10.2210/pdb4ODH/pdb). The V and the C domain, as well as the variable heavy (V ) and constant L L H heavy (C ) domains are labeled. The LC is marked green, the heavy chain is displayed in gray. Red sphere: residue homologous to the FOR001 glycosylation site. b Crystal structure of a LC dimer encompassing an IGLV1-51*02 GL segment (PDB 5MUD 10.2210/pdb5MUD/pdb). One LC in the dimer is marked green, the other in gray. Red sphere: as in (a). c Fibril-formation kinetics of refolded FOR001 fibril protein as obtained from real-time measurements of the ThT fluorescence intensity. Blue: deglycosylated; red: glycosylated protein. d Coomassie-stained denaturing-protein electrophoresis gels of samples to estimate the proteolytic stability of ex vivo FOR001 fibrils and fibrils formed in vitro from deglycosylated and glycosylated FOR001 protein. Each gel was replicated three times. M: marker. e Densitometric quantification of the fibril protein band (n = 3) of ex vivo fibrils (gray) and in vitro fibrils from deglycosylated (blue) and glycosylated FOR001 protein (red) after digestion with proteinase K for different periods of time. The band intensity of the sample before proteinase K addition was set to 100%. Based on a one-tailed Welch t-test, the amounts of glycosylated and deglycoslated fibril proteins differ from one another with a p-value of 0.032 and 0.049 at time points 0 min and 1 min, respectively. Error bars represent the standard deviation. in vitro, we purified and refolded FOR001 fibril protein from the four known ex vivo amyloid fibril structures from systemic AL heart. The glycosylation was removed from a fraction of the amyloidosis. Each of these structures is different (Supplementary refolded protein. Both the glycosylated and the deglycosylated Fig. 7). The observed differences include the precise position of protein variants were able to form fibrils in vitro, as indicated by the β-strands and disordered regions within the fibril protein time-resolved fibrillation measurements with the amyloid- sequence, the presence or absence of internal cavities, the topo- binding dye thioflavin T (ThT) (Fig. 5c). The deglycosylated logical arrangement of the secondary structural elements, the FOR001 fibril protein aggregated much faster under these con- organization of the network of chemical interactions that stabilize ditions than the glycosylated fibril protein (Fig. 5c), demon- the misfolded conformational state, and the involvement of strating that a glycosylation retards rather than accelerates fibril PTMs. However, the fibrils share structural features such as a formation in vitro. We further investigated the proteolytic sta- fibril core that is mainly derived from the LC V domain, the bility of the glycosylated and deglycosylated fibril samples presence of an intramolecular disulfide bond, and the tertiary (Fig. 5d). Both in vitro fibril samples are substantially less stable structures of the fibril proteins are substantially different from a to proteolysis with proteinase K than the ex vivo fibrils, although natively folded LC. the glycosylated in vitro fibrils seem slightly more stable to pro- In the FOR001 fibril, we identified three PTMs: a disulfide teolysis than the deglycosylated in vitro fibrils (Fig. 5e). bond (Figs. 2d and 3c), a Pyro-Glu modification that affects only a fraction of the fibril proteins (Supplementary Figs. 4 and 5a, b), and a carbohydrate moiety that is linked by N-glycosylation Discussion (Supplementary Fig. 5d). The disulfide bond corresponds to the In this study, we obtained the cryo-EM structure of a glycosylated canonical disulfide of a natively folded LC V domain . It occurs λ1 AL amyloid fibril that was extracted from the heart tissue of in all previously described cryo-EM structures of ex vivo AL patient FOR001. Adding this structure to the previously pub- 21–23 21–23 amyloid fibrils , where it connects two chain segments that lished ex vivo AL amyloid fibril structures , there are now 6 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications m molecular mass (kDa) 0 min 1min 2 min 5 min 0 min 1 min 2 min 5 min 0 min 1 min 2 min 5 min fibril protein (%) relative ThT flourescence (%) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE present an orientation that is switched by 180° relative to the fibril protein structures or fibril morphologies are associated with 21 44–46 native state . This disulfide has profound effects on the mis- different disease variants . Ex vivo fibril morphologies are folding and on the fibril structure, as it restricts the conforma- often different from fibrils formed in vitro from the same 20,23,44,47–49 tional freedom of the fibril protein and leads to the requirement proteins and more protease-stable than their in vi- 47,49 of a rotational switch around the disulfide bond as a key mole- tro formed counterparts . In particular, the latter observations cular event in the misfolding of amyloidogenic LCs . led to the hypothesis that pathogenic amyloid fibril structures 24,25 Pyro-Glu residues were previously observed in some but may have been selected inside the body by their high proteolytic 21–23 not all AL fibril proteins . They also occur in N-terminally resistance and ability to escape endogenous clearance 33 49,50 truncated Alzheimer’sAβ peptide and may accelerate the systems . The present observations further support this view fibrillation of this peptide in vitro . In the FOR001 fibril, a sig- as they imply that the disulfide formation, the glycosylation, and nificant fraction of the fibril protein lacks the Pyro-Glu mod- the mutational changes inserted during somatic hypermutation ification (Supplementary Figs. 4 and 5a, b), and the modification helped to shape the specific fibril morphology that is associated affects a disordered region of the fibril structure (Fig. 3a). While with patient FOR001 and which shows a high protease stability we cannot exclude that Pryo-Glu can affect the rate of aggrega- (Fig. 5d, e). tion, there is, based on our data, no strong evidence to suggest In summary, we have shown that in AL amyloidosis, fibril that Pyro-Glu is a major driver for the generation of the specific deposits can interfere with the physiological function of heart fibril morphology observed in patient FOR001. tissue by disrupting its ordered architecture. PTMs and the spe- The third PTM is an N-glycosylation. The Asn–Xaa–Ser/Thr cific mutational changes that characterize amyloidogenic LCs N-glycosylation motif occurs more frequently in AL LC sequen- facilitate the formation of patient-specific fibril morphologies that 35,36 ces than in normal LCs , and glycosylated LCs are over- are able to survive under the otherwise hostile and proteolytic represented in systemic AL amyloidosis patients , which conditions inside the body. The resulting fibrils are thus able to indicates that LC glycosylation is important for systemic AL accumulate, proliferate, and cause damage to the surrounding amyloidosis. However, the biophysical basis of this effect tissue. The differences in the available cryo-EM structures of remained obscure. Several studies demonstrated that glycosyla- fibrils from AL amyloidosis patients underline the patient-specific 37,38 tion can stabilize the native fold of globular proteins , reduce nature of the disease and make clear the need for investigating AL the protein conformational dynamics , or increase the solubility fibril structures from many patient cases in order to identify 37,38 of proteins , indicating that a glycosylation should render a common principles of AL amyloid formation. protein less amyloidogenic. Indeed, deglycosylated FOR001 fibril protein aggregates faster than glycosylated protein, as demon- Methods strated here by in vitro fibrillation measurements with ThT Source of the fibril-containing tissue. The fibrils were extracted from the (Fig. 5c). explanted heart of a male patient (FOR001), who suffered from systemic AL amyloidosis with cardiac involvement and underwent cardiac surgery at the age of A potentially more important effect of glycosylation could be 51 years. The underlying condition was a monoclonal gammopathy. The patient that it masks the amyloid deposits within the tissue, preventing was treated within the heart-transplant program of the University Hospital Hei- their clearance by body-own mechanisms . A similar amyloid- delberg. The explanted heart tissue was stored at −80 °C. The study was approved masking effect is known for serum amyloid P component that is by the ethical committees of the University of Heidelberg (123/2006) and of Ulm University (203/18). Informed consent was obtained from the patient for the also glycosylated. The carbohydrate of the FOR001 protein is analysis of the amyloid deposits. located on the surface of the fibril structure (Fig. 2b, d) and of the natively folded LC, as indicated by analysis of the crystal structure Visualization of amyloid fibril deposits in heart tissue using STEM and SEM. of a homologous LC (Fig. 5a, b). Hence, the carbohydrate covers Frozen pieces of FOR001 heart tissue (~1 mm )were fixed in a solution of 0.1% (w/v) part of the surface of the fibrils or its biological precursors, glutaraldehyde, 4% (w/v) paraformaldehyde, and 1% (w/v) saccharose in 0.1 M sodium consistent with a protective effect and with observations that phosphate buffer (pH 7.3) overnight at 4 °C. The further sample preparation for STEM glycosylated in vitro fibrils from FOR001 fibril protein might be and SEM was based on a protocol from a previous publication .In brief,the fixed tissue pieces were cut into ~200-μm slices with a scalpel and then high-pressure frozen slightly more stable to proteinase K digestion than in vitro formed with a Compact 01 high-pressure freezing device (Engineering Office M. Wohlwend) fibrils from deglycosylated protein. However, both in vitro fibrils and freeze-substituted in a medium consisting of 0.1% (w/v) uranyl acetate, 0.2% (w/v) are much less stable to proteinase K digestion than the ex vivo osmium tetroxide, and 5% (v/v) water in acetone using an AFS2 freeze-substitution FOR001 fibrils (Fig. 5d, e). device (Leica Microsystems) with which the temperature was raised from −90 °C to room temperature over a period of 19 h. Afterward, the tissue pieces were embedded in The most profound effect of glycosylation suggested by our epoxy resin (Sigma-Aldrich) starting with a mixture of 30% (v/v) resin in acetone for data is that it helps to define the fibril protein fold in this patient. 1 h followed by 60% (v/v) resin in acetone for 3 h and 100% resin overnight and The fibril protein tertiary structure in the glycosylated λ1 fibril of polymerization in fresh 100% resin by incubation at 60 °C for 48 h. For SEM imaging, FOR001 is significantly different from a previously described, 200-nm thin sections were cut from the polymerized samples using the ultra- microtome Ultracut (Leica Microsystems) equipped with a 45° diamond knife (Dia- nonglycosylated λ1 fibril . The previous structure, and all known tome). Sections were mounted on glow-discharged silicon wafers and stained with a structures of nonglycosylated AL amyloid fibril, is defined by a 0.3% (w/v) lead citrate solution in water for 1 min, washed with distilled water, dried, juxtaposition of the N- and C-terminal ends of the fibril protein and imaged in a Hitachi S-5200 field emission scanning electron microscope, detecting (Supplementary Fig. 7). In the FOR001 fibril, which carries the the secondary electron signal at 5 kV in analysis mode. For STEM, 300-nm thin glycosylation in the N-terminal end of the fibril protein, no such sections were cut with a 45° diamond knife and processed in a similar way as described in a previous publication . In brief, sections were mounted on copper grids with 200 juxtaposition is observed and the N- and C-terminal ends of the parallel grid bars (Plano). After attachment of 15-nm colloidal gold fiducials (Aurion), polypeptide chain are spread apart in the tertiary structure. sections were coated with carbon by electron-beam evaporation in a Baf 300 (BalTec). Hence, the glycosylation may help to favor the FOR001 fibril Tomograms were acquired with a STEM JEM-2100F (JEOL) operated at 200 kV. Tilt protein fold by preventing the association of the N- and series were acquired from −72° to +72° with a 1.5° increment using the bright-field detector. The pixel size was 1.395 nm. Tilt series were reconstructed to tomograms by C-terminal ends of the fibril protein. weighted back projection using an emulated simultaneous iterative-reconstruction Several previous observations have suggested that the fibril technique-like filter (20 iterations) and segmentation of fibrils and cell membranes was protein fold and the fibril morphology are crucial for the 53 performed with the IMOD software package , version 4.9.0. pathological process of an amyloid disease. Consistent fibril protein folds and fibril morphologies can be found in different Measurement of the persistence length. A total of 197 amyloid fibrils in heart patients or animals that are affected by the same disease variant tissue were traced in the STEM tomograms and were then available as polygonal 42–44 and allelic variant of the fibril precursor protein . Different chains. Based on this representation, the squared end-to-end distance R, i.e., the NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 Cartesian plane distance between the starting and end points, and the contour particles were subjected to a 3D classification with the initial model as a reference. length L, i.e., the sum of the length of all line segments of the polygonal chain, were This treatment resulted in a map with fibril-like features. Using this map as a computed for each fibril. The persistence length P of the fibrils was then deter- reference, the particle set selected from 2D classification was subjected to several mined by regression analysis using the formula rounds of 3D classification, 3D refinement, and post-processing. The resolution of the resulting map was estimated to be 3.6 Å. To obtain a more homogeneous set of P L P particles, fibrils were picked more selectively, yielding 43,308 particles with a box R ¼ 2PL 1  1  e ð1Þ size of ~270 Å. Using the previously obtained post-processed map as a reference 54 (with adjusted box size), subsequent steps of 3D classification, 3D refinement, and according to Kollmer et al. and using the curve-fitting tool in Matlab (MATLAB post-processing yielded a map at a resolution of 3.4 Å. All manually picked par- 2019, The MathWorks). P is contained in the confidence interval from 0.66 μmto ticles were retained in this reconstruction. Further improvement of map resolution 0.81 μm in 95% of all cases. The bending rigidity B is computed based on P was accomplished through Bayesian polishing, resulting in a final map resolution according to the formula of 3.1 Å, based on the value of the FSC curve for two independently refined half- B ¼ k TP ð2Þ maps at 0.143. An estimated map-sharpening B-factor of −67.383 Å was applied. The map had a twist of −1.45566° and a rise of 4.76311 Å. These values agree with where k and T denote the Boltzmann constant and the temperature (300 K), the twist value calculated from the measured crossover distances on the cryo-EM 21,55 respectively . micrographs, and with the rise value measured from the micrograph power spectra. Fibril extraction from FOR001 heart tissue. Fibrils were extracted from FOR001 56 Model building and refinement. The software Coot v0.8.9 was used to manually heart tissue using a previously established protocol . In brief, 250 mg of heart build the protein model de novo. First, the 3D map was traced and a poly-L-Ala tissue was diced with a scalpel and washed five times with Tris Calcium Buffer chain created representing the protein backbone. The residues of this chain were (TCB) [20 mM Tris, 138 mM NaCl, 2 mM CaCl , and 0.1% (w/v) NaN , pH 8.0]. 2 3 then mutated to the FOR001 fibril protein sequence as determined by MS. This Each washing step consisted of the addition of 0.5 mL of ice-cold TCB to the pellet, initial model was then further improved by subsequent rounds of manual and the homogenization of the tissue with a Kontes Pellet Pestle, and the separation of automated refinement (phenix.real_space_refine) as implemented in Phenix the tissue from the supernatant by centrifugation for 5 min at 3100 × g and 4 °C, v1.16 . Noncrystallographic symmetry and secondary-structure restraints were followed by the removal of the supernatant. The tissue pellet after the fifth washing imposed. Model-based automated sharpening of the map (phenix.auto_sharpen) step was resuspended in 1 mL of TCB containing 5 mg/mL Clostridium histolyti- yielded an improved map, which was used to further refine the model. The quality cum collagenase (Sigma-Aldrich) and one tablet of complete ethylenediaminete- of the model was assessed using the MolProbity -generated validation report. traacetic acid (EDTA)-free protease-inhibitor cocktail (Roche) per 7 mL of TCB. Modeling parameters are listed in Supplementary Table 1. After an overnight incubation of the digest at 37 °C in a horizontal orbital shaker (750 rpm), the sample was centrifuged for 30 min at 3100 × g and 4 °C. The pellet was subjected to ten washing steps which were performed in the same manner as Protein sequence determination by electrospray-ionization MS. About 2 μgof the TCB washing steps described above, except that 0.5 mL of Tris EDTA buffer refolded, lyophilized and glycosylated fibril protein was resuspended in 15 μLof [20 mM Tris, 140 mM NaCl, 10 mM EDTA, and 0.1% (w/v) NaN , pH 8.0] was buffer [280 mM Tris/HCl pH 6.8, 9% (w/v) sodium dodecyl sulfate (SDS), 33.3% used instead of TCB. The pellet from the last wash was resuspended in 0.5 mL of (w/v) glycerol, and 100 mM dithiothreitol] and processed by denaturing protein gel ice-cold distilled water, mixed with a pipette, and centrifuged for 5 min at 3100 × g electrophoresis. Afterward, the gel band of the fibril protein was cut out and and 4 °C. The fibril-containing supernatant was retained and the pellet was sub- washed by a 10-min incubation in the respective protease buffer (see below), and mitted to nine more cycles of resuspension in water and centrifugation. The subsequently, in a mixture of 50% (v/v) protease buffer and 50% (v/v) ACN for supernatants were retained to check for the presence of fibrils. 10 min. These incubation steps were repeated twice, followed by vacuum drying. Dried gel slices were reduced with 5 mM dithiothreitol (AppliChem) in 50 mM ammonium bicarbonate, pH 8.0, for 20 min at room temperature and subsequently Platinum side-shadowing and TEM. The handedness of the fibrils was deter- alkylated with 55 mM iodoacetamide (Sigma-Aldrich) in 10 mM ammonium mined by platinum side-shadowing and TEM. Formvar and carbon-coated 200 bicarbonate for 20 min at 37 °C. The gel slices were placed in five different protease mesh copper grids (Plano) were glow discharged for 40 s at 40 mA using a PELCO solutions (trypsin in 50 mM ammonium bicarbonate, pH 8.0; LysC in 50 mM easiGlow glow-discharge cleaning system (Ted Pella). About 15 μL of the fibril ammonium bicarbonate, pH 8.0; elastase in 50 mM Tris/HCl, pH 9.0; chymo- solution were applied to the grid and incubated for 30 s at room temperature. The grid was blotted using filter paper (Whatman) to remove excess fluid. The grid was trypsin in 50 mM Tris/HCl buffer, pH 8.0, 10 mM CaCl ; pepsin 40 mM HCl, pH 1.5). Each protease was used at 0.33 ng/μL concentration and digestion was carried washed three times with 10 μL of distilled water and dried at room temperature. out overnight at 37 °C (except for chymotrypsin at 25 °C). The resulting peptides Platinum was evaporated at an angle of 30° onto the grid to form a 1-nm-thick were released from the gel slices in two steps: the first step was to add 20 μLof a layer by use of a Balzers BAF 300 coating device. Grids were imaged using a JEM- solution containing 50% (v/v) ACN and 0.1% (v/v) TFA; the second step was an 1400 TEM (JEOL) that was operated at an acceleration voltage of 120 kV. The incubation in an ultrasonic bath (Bandelin Sonorex Super 10 P) at 100% intensity images were recorded with an F216 camera (TVIPS). for 10 min each. ACN was evaporated and samples were filled to 15 μL with 0.1% TFA (v/v). Cryo-EM sample preparation and data collection.C-flat 1.2/1.3 400-mesh holey Samples were separated by liquid chromatography using a U3000 RSLCnano carbon-coated grids (Science Services) were glow-discharged at 40 mA for 40 s (Thermo Fisher Scientific) online coupled to the mass spectrometer with an using a PELCO easiGlow glow-discharge cleaning system (Ted Pella). Conditions Acclaim PepMap analytical column (75 μm × 500 mm, 2 μm, 100-Å pore size, of grid preparation were optimized with the help of a Vitrobot Mark 3 (Thermo Thermo Fisher Scientific) in combination with a C18 μ-precolumn Fisher Scientific) and checked in a 200-kV JEM 2100 F transmission electron (0.3 mm × 5 mm, PepMap, Dionex LC Packings, Thermo Fisher Scientific). First, microscope (JEOL) that was equipped with a DE12 detector (Direct Electron). The samples were washed with 0.1% (v/v) TFA for 5 min at a flow rate of 30 μL/min. grids for data collection were prepared by application of 3.5 μLof fibril solution to Subsequent separation was carried out employing a flow rate of 250 nL/min using a a grid, incubation for 30 s at >95% humidity, both-side blotting using filter paper gradient consisting of solvent A [0.1% (v/v) formic acid] and solvent B [86% (v/v) (Whatman), and plunging into liquid ethane (~103 K). The data set was recorded ACN, 0.1% (v/v) formic acid]. The main column was initially equilibrated in a with a Titan Krios transmission electron microscope (Thermo Fisher Scientific) at mixture containing 5% (v/v) solvent B and 95% (v/v) solvent A. For elution, the 300 kV and applying a Gatan imaging filter with a 20-eV slit. The images were percentage of solvent B was raised from 5 to 15% over a period of 10 min, followed recorded with a K2-Summit detector (Gatan) in counting mode. The software by an increase from 15 to 40% over 20 min. Fractions from the main column SerialEM v3.7 was used for data collection. In total, 3033 micrographs were col- directly eluted into the ionization module and were further analyzed by MS. lected from a single grid. See Supplementary Table 1 for further details. Global Samples were measured using an LTQ Orbitrap Velos Pro system (Thermo fibril parameters, such as width and crossover distance, were measured using Fiji Fisher Scientific). The mass spectrometer was equipped with a nanoelectrospray v1.52 . The proportion of fibrils showing the reconstructed morphology was ion source and distal-coated SilicaTips (FS360-20-10-D, New Objective). The determined by analyzing all fibrils (length at least 200 nm) in 100 micrographs. instrument was externally calibrated using standard compounds (LTQ Velos ESI Positive Ion Calibration Solution, Pierce, Thermo Scientific). The system was Reconstruction of the 3D map. Motion correction and dose-weighting was car- operated using the following parameters: spray voltage, 1.5 kV; capillary ried out with MotionCor2 . For predicting, refining, and correction of the con- temperature, 250 °C; S-lens radio-frequency level, 68.9%. The software XCalibur trast transfer function, Gctf v1.06 was used. Helical reconstruction was 2.2 SP1.48 (Thermo Fisher Scientific) was used for data-dependent MS/MS performed using Relion 3.0 .As a first step, 279,338 particles were picked analyses. Full scans ranging from mass-to-charge ratio (m/z) 370–1700 were manually with a box size of ~312 Å and an interbox distance of 34.6 Å (~11%). acquired in the Orbitrap at a resolution of 30,000 (at m/z 400) with automatic gain After a reference-free 2D classification using 279 classes and a regularization value control enabled and set to 10 ions and a maximum fill time of 500 ms. Collision- of T = 2 in the first run, another 2D classification was performed with 50 classes. induced dissociation was employed as the fragmentation method on individual 2D classes were selected based on the visibility of a z-axial repeat at ~4.8 Å and bad sample sets. Per survey scan, 10 ions were selected. Single charged ions were classes were excluded. An initial 3D model was generated using Relion’s initial- rejected and the m/z peaks of fragmented single-charged ions were excluded from model job. To generate a better reference for 3D classification, fibrils with visible fragmentation for 60 s. In the linear ion trap, the automatic gain control was set to crossovers were picked from 28 micrographs, resulting in 244 particles. These 10,000 ions and a maximum fill time of 100 ms. For MS/MS fragmentation, a 8 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE normalized collision energy of 35% with an ‘activation q’ of 0.25 and an activation [500 mM sodium phosphate, pH 7.5] (New England Biolabs), 1/2 V water, and Prot time of 30 ms was used. The resulting fragments were analyzed using the linear 1/10 V PNGase F in 20 mM Tris/HCl, 50 mM NaCl, and 5 mM EDTA, pH 7.5 Prot ion-trap part at rapid scan speeds. Subsequent detection of fragmentation spectra (New England Biolabs), and incubated for 1 h at 37 °C. For O-deglycosylation, the was performed in the Orbitrap mass analyzer at a resolution of 7500. heated sample was mixed with 1/5 V of a 10% (v/v) Nonident P40 solution Prot For de novo sequencing, the Peaks AB Software (Bioinformatics Solutions) was (New England Biolabs), 1/5 V 10 × Glycobuffer 2 (New England Biolabs), 3/10 Prot used. The resulting sequence was then used as a target for further analyses using V water, 1/10 V O-glycosidase in 20 mM Tris/HCl, 50 mM NaCl, and 1 mM Prot Prot the Peaks X software suite (Bioinformatics Solutions) in order to confirm the de EDTA, pH 7.5 (New England Biolabs), and 1/5 V neuraminidase in 20 mM Prot novo sequence. For all analyses, the mass accuracy was set to 10 ppm on intact Tris/HCl, 50 mM NaCl, and 5 mM EDTA, pH 7.5 (New England Biolabs), and peptide masses and 0.5 Da. Various PTMs were considered, including the incubated for 10 min at 37 °C. The deglycosylation was checked by denaturing deamidation of Asn or Gln residues, pyroglutamate modifications (Gln), oxidation protein-gel electrophoresis. N- and O-glycosylated fetuin protein, which we pur- of Met, as well as carbamidomethylated Cys as a result of the alkylation. Ile/Leu as chased as a 10 mg/mL solution from New England Biolabs, was used as a control well as Gln/Lys have the same molecular weights and could not be uniquely substance. For the experiments reported in Fig. 5, the lyophilized, refolded FOR001 determined. These residues were assigned based on homology considerations. protein was dissolved in water at 1 mg/mL concentration and denatured—without SDS and dithiothreitol—for 10 min at 100 °C, as the SDS in the denaturing buffer was found to interfere with the subsequent reversed-phase chromatography. MS analysis of the total fibril protein mass. Lyophilized FOR001 fibril protein was resuspended in Glycoprotein Denaturing Buffer (New England Biolabs) and Protein-concentration measurement based on the intrinsic absorbance at subsequently it was deglycosylated without the heating step of the standard pro- 280 nm. About 40 μL of protein solution was mixed with 160 μL of 7.5 M gua- tocol. The deglycosylated protein was diluted with 0.1% TFA (v/v) to obtain a nidine hydrochloride (Carl Roth) in 25 mM sodium phosphate buffer, pH 6.5. The concentration of 66 μg/mL. It was applied with a flow rate of 10 μl/min onto a absorbance was measured at 280 nm in a Lambda Bio+ ultraviolet/visible (Perki- PepSwift trap column (200 μm × 5 mm, Thermo Fisher Scientific) in combination nElmer) spectrometer using a Quartz Suprasil R Ultra-Micro cuvette (Hellma). The with a monolithic ProSwift RP-4H analytical column (100 μm × 50 cm, Thermo protein concentration was determined based on the Lambert–Beer law using a Fisher Scientific), which was connected to a U3000 RSLCnano (Thermo Fisher −1 −1 theoretic molar extinction coefficient of 16,740 M cm for the FOR001 fibril Scientific) that was coupled to the mass spectrometer. The fibril protein was eluted protein according to the method of Gill and von Hippel . using a gradient of solvent B [86% (v/v) ACN, 0.1% (v/v) formic acid] and solvent A [0.1% (v/v) formic acid] with a flow rate of 1 μL/min. The gradient started with an increase of 5–55% solvent B over a period of 75 min, followed by an increase Fibril-formation kinetics measurements using ThT. Refolded and lyophilized from 55 to 95% over 15 min. The concentration of 95% solvent B stayed constant FOR001 fibril protein (glycoslylated or deglycosylated) was dissolved at 2 mg/mL for 3 minutes with a subsequent reduction from 95 to 5% solvent B over 9 min. concentration in water. Aggregation kinetics measurements were set up in PF 96- Fractions from the ProSwift RP-4H analytical column directly eluted into the well F-bottom black microplates (Greiner Bio-One International). Each well was ionization module and were further analyzed by MS. Samples were measured using filled with 100 μL of sample, containing 0.4 mg/mL glycosylated or deglycosylated an LTQ Orbitrap Elite system (Thermo Fisher Scientific). The mass spectrometer fibril protein, 20 μM ThT and 10 mM sodium acetate, 10 mM boric acid, 10 mM was equipped with a nanoelectrospray ion source and distal-coated SilicaTips sodium citrate, pH 4.0, and 150 mM NaCl. The plates were sealed with Rotilabo- (FS360-20-10-D, New Objective). The instrument was externally calibrated using sealing film (Carl Roth) and incubated at 37 °C in FLUOstar Omega (BMG Lab- standard compounds (LTQ Velos ESI Positive Ion Calibration Solution, Pierce, tech) for 72 h. During incubation, the plates were agitated by orbital shaking at Thermo Scientific) and operated using the following parameters: spray voltage, 300 rpm, which was paused during measurement. The fluorescence emission 1.5 kV; capillary temperature, 250 °C; S-lens radio-frequency level, 68.9%. The intensity at 490 nm was recorded every 30 min upon excitation at 450 nm. software XCalibur 2.2 SP1.48 (Thermo Fisher Scientific) was used for data- dependent MS/MS analyses. Full scans ranging from mass-to-charge ratio (m/z) Proteinase K digestion of amyloid fibrils. Aliquots of solutions containing ex vivo 370–1700 were acquired in the Orbitrap at a resolution of 30,000 (at m/z 400) with FOR001 amyloid fibrils or in vitro formed fibrils from glycosylated or deglycosylated automatic gain control enabled and set to 10 ions and a maximum fill time of FOR001 fibril protein (from the ThT kinetic experiment) were mixed with water and 500 ms. The raw data were deconvoluted by the MASH Explorer using default a 10 × buffer stock [200 mM Tris, pH 8.0, 1.4 M NaCl, 20 mM CaCl , and 1% (w/v) settings and the Quick Deconvolution feature. All calculated monoisotopic masses NaN ] to reach a total volume of 60 μL containing 0.2 mg/mL protein in 1 × buffer. A with a score equal to or above 94% resulting from initial m/z peaks with 5 charges first aliquot (10 μL) was withdrawn from this solution and retained for gel electro- or more were considered as correct and are shown in Supplementary Fig. 5b, c. The phoresis as the control sample without protease. The remaining 50 μL of the protein deconvoluted mass peaks were further assigned to protein species by using the solution were mixed with 1 μL of a 2 mg/mL proteinase K solution (Thermo Fisher software mMass considering a tolerance of 0.1 Da and a peak charge of 0. Scientific). Immediately afterward, a second aliquot (10 μL) was removed (0-min Sequence modifications were set as follows: pyroglutamylation at Gln1 was set as sample). The remaining solution was incubated at 37 °C in a heating block and variable, whereas the disulfide bond between Cys22 and Cys89 was set as fixed. further aliquots (10 μL) were withdrawn after 1 min, 2 min, and 5 min. As soon as an aliquot was withdrawn, it was mixed with 1 μL of 200 mM phenylmethylsulfonyl fluoride (PMSF) (Carl Roth) in methanol, incubated for 1 min at room temperature, Refolding of the FOR001 fibril protein. Solid guanidine hydrochloride was added and flash-frozen in liquid nitrogen. After the experiment, all aliquots were brought to to a sample of ex vivo FOR001 fibrils to reach a final concentration of 6 M followed room temperature and analyzed by denaturing protein-gel electrophoresis. The by an overnight incubation at room temperature to disaggregate the fibrils. The resulting protein bands (72 × 150 pixels) were densitometrically quantified using the protein was refolded by dialysis (molecular weight cutoff 3.5 kDa, Spectra/Por 6 program Fiji v1.52 . The intensity of the fibril protein band without proteinase K Dialysis Membrane Pre-wetted RC Tubing, Spectrum Labs) against 20 mM Tris was set to 100%, and an equally sized area on the gel without protein at 0%. buffer, pH 8.0, for 24 h at 4 °C. The protein was purified by anion-exchange chromatography with Q-SepharoseFF medium (10 mL, Cytiva) in an XK 16/20 column (Cytiva) with a slope gradient from 0% to 100% elution buffer [20 mM Tris Denaturing protein gel electrophoresis. Samples from the deglycosylation buffer, 1 M NaCl, pH 8.0] over 20 column volumes (CVs). The fibril protein- experiment and proteolytic stability measurement (10-μL volume for deglycosy- containing fractions, as identified by protein-gel electrophoresis, were purified lation, 11-μL volume for proteolytic stability, including 1-μL PMSF) were mixed further with a Resource 15 RPC column (3 mL, Cytiva) that was equilibrated in with 2 μL of 10 × NuPAGE reducing agent (Thermo Fisher Scientific), 5 μLof solvent A [0.1% (v/v) trifluoroacetic acid (TFA) in water]. The protein was eluted 4 × NuPAGE LDS sample buffer (Thermo Fisher Scientific), and water to generate through a slope gradient from 0 to 58% solvent B [86% (v/v) acetonitrile (ACN), a sample with a total volume of 20 μL. The solution was heated at 95 °C for 10 min 0.1% (v/v) TFA] over 20 CVs, followed by second gradient from 58 to 100% solvent and applied onto a 4–12% NuPAGE Bis-Tris gel (Thermo Fisher Scientific), B over 4 CVs to remove other bound proteins. Fractions were collected and the operated in NuPAGE MES SDS running buffer (Thermo Fisher Scientific). Blue- fractions containing the FOR001 fibril protein were identified by protein gel Easy Prestained (Genetics) was used as a marker. The gel was stained in a solution electrophoresis, pooled, and lyophilized. containing 30% (v/v) ethanol, 10% (v/v) acetic acid, and 0.25% (w/v) Coomassie brilliant blue and destained with a solution containing 20% (v/v) ethanol and 10% (v/v) acetic acid. Deglycosylation of the refolded FOR001 fibril protein. For the experiment shown in Supplementary Fig. 5a, the lyophilized, refolded FOR001 fibril protein was dissolved in water at approximately 2 mg/mL concentration. The exact protein Analysis of GL segments and mutations. The FOR001 amino acid sequence was concentration was determined by the intrinsic protein absorbance at 280 nm. This used to search the IgBLAST database (https://www.ncbi.nlm.nih.gov/igblast/), solution was mixed with water and SDS-containing 10 × glycoprotein-denaturing which returned the GL segment IGLV1-51*02. The protein sequence of the GL buffer [5% (w/v) SDS, 400 mM dithiolthreitol] (New England Biolabs) to generate a segment was retrieved from the VBase2 database (http://www.vbase2.org/). The final FOR001 fibril protein solution with the volume V containing a protein FOR001 J segment was compared with the five functional IGLJ GL segments in the Prot concentration of 1 mg/mL and 1 x Glycoprotein Denaturing Buffer (New England GenBank database (https://www.ncbi.nlm.nih.gov/genbank/). It matched both Biolabs). The protein was heated for 10 min at 100 °C to partially denature the IGLJ2 (Gene ID: 28832) and IGLJ3 (GeneID: 28831) and no unique source could be protein before it was cooled to room temperature. For N-deglycosylation, the determined. Nor could we identify the GL precursor of the C segment as most of heated sample was mixed with 1/5 V of a 10% (v/v) solution of the detergent the C domain is missing in the FOR001 fibril protein. The residues Leu97–Ala98 Prot L Nonident P40 in water (New England Biolabs), 1/5 V 10 × Glycobuffer 2 constitute the variable V/J junctional region. CDRs were determined with abYsis Prot NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 (http://www.abysis.org/abysis/) based on the Kabat definition . In this paper, all 9. Abraham, R. S. et al. Immunoglobulin light chain variable (V) region genes mutations are represented in the direction GL to fibril protein. influence clinical presentation and outcome in light chain–associated amyloidosis (AL). Blood 101, 3801–3807 (2003). 10. Kourelis, T. V. et al. Clarifying immunoglobulin gene usage in systemic and Computation of the aggregation score. To compute the aggregation score, we localized immunoglobulin light-chain amyloidosis by mass spectrometry. 69 70 71 72 used TANGO version 2.1 , WALTZ , FoldAmyloid , Aggrescan , and 73 Blood 129, 299–306 (2017). PASTA2.0 . These programs calculate a residue-specific aggregation potential. 11. Perfetti, V. et al. The repertoire of lambda light chains causing predominant The following settings were chosen for each program: TANGO: temperature: amyloid heart involvement and identification of a preferentially involved 309.15 K, ionic strength: 0.02 M, and concentration: 1 M and pH 7.0; WALTZ: the germline gene, IGLV1-44. Blood 119, 144–150 (2012). threshold was set to high sensitivity and the pH to 7.0; FoldAmyloid: scale: triple 12. Hurle, M. R., Helms, L. R., Li, L. I. N., Chan, W. & Wetzel, R. A role for hybrid, averaging frame: 5; Aggrescan: default settings; PASTA: 90% specificity, top destabilizing amino acid replacements in light-chain amyloidosis. Proc. Natl. pairing energies: 22. Residues with a high aggregation potential are defined as Acad. Sci. USA 91, 5446–5450 (1994). follows: TANGO: β-sheet aggregation values above 0.0; WALTZ: total sequence 13. Wall, J. et al. Thermodynamic instability of human lambda 6 light chains: score above 0.0; Foldamyloid: five successive amino acids with a triple-hybrid threshold above 0.062; Aggrescan: aggregation-propensity values above −0.02; Correlation with fibrillogenicity Biochemistry. Biochem. 38, 14101–14108 (1999). PASTA: PASTA energy units below −2.8. An aggregation score of 0 means that none of the programs identifies a high aggregation score for a given residue. An 14. Oberti, L. et al. Concurrent structural and biophysical traits link with aggregation score of 5 means that all five programs predict a high aggregation score immunoglobulin light chains amyloid propensity. Sci. Rep. 7,1–11 (2017). for that residue. 15. Kazman, P. et al. Fatal amyloid formation in a patient’s antibody light chain is caused by a single point mutation. eLife 9, e52300 (2020). 16. Rottenaicher, G. J. et al. Molecular mechanism of amyloidogenic mutations in Protein structure representation. The images of the density map and protein 74 hypervariable regions of antibody light chains. J. Biol. Chem. 296, 100334 model were created with the software UCSF Chimera v1.14 . Hydrogen bonds (2021). were defined according to the criteria of both UCSF Chimera v.1.14 and the 61 17. Blancas-Mejía, L. M. et al. Kinetic control in protein folding for light chain software Coot v0.8.9 . The β-sheets were defined as a minimum of two residues amyloidosis and the differential effects of somatic mutations. J. Mol. Biol. 426, having φ/ψ angles in the β-sheet region of the Ramachandran plot and at least one- 347–361 (2014). backbone hydrogen bond. 18. Piehl, D. W., Blancas-Mejía, L. M., Ramirez-Alvarado, M. & Rienstra, C. M. Solid-state NMR chemical shift assignments for AL-09 V L immunoglobulin Sample statistics. In this paper errors report the standard deviation. In Fig. 5e, a light chain fibrils. Biomol. NMR Assign. 11,45–50 (2017). one-tailed Welch t-test was used. 19. Pradhan, T. et al. Seeded fibrils of the germline variant of human λ-III immunoglobulin light chain FOR005 have a similar core as patient fibrils with Reporting summary. Further information on research design is available in the Nature reduced stability. J. Biol. Chem. 295, 18474–18484 (2020). Research Reporting Summary linked to this article. 20. Annamalai, K. et al. Common fibril structures imply systemically conserved protein misfolding pathways in vivo. Angew. Chem. 129, 7618–7622 (2017). 21. Radamaker, L. et al. Cryo-EM structure of a light chain-derived amyloid Data availability fibril from a patient with systemic AL amyloidosis. Nat. Commun. 10.1,1–8 The datasets used during the current study are available from public repositories and/or (2019). from the corresponding author on reasonable request. The cryo-EM map of the FOR001 22. Swuec, P. et al. Cryo-EM structure of cardiac amyloid fibrils from an fibrils was deposited in the Electron Microscopy Data Bank (https://www.ebi.ac.uk/pdbe/ immunoglobulin light chain AL amyloidosis patient. Nat. Commun. 10.1,1–9 emdb/) with the accession code EMD-12570. The coordinates of the corresponding (2019). atomic model were deposited in the PDB (https://www.rcsb.org/) under the accession 23. Radamaker, L. et al. Cryo-EM reveals structural breaks in a patient-derived code 7NSL. The cryo-EM data of the FOR001 fibrils were deposited on EMPIAR (https:// amyloid fibril from systemic AL amyloidosis. Nat. Commun. 12,1–10 (2021). www.ebi.ac.uk/pdbe/emdb/empiar/) with the accession code EMPIAR-10730. The 24. Yazaki, M., Liepnieks, J. J., Callaghan, J., Connolly, C. E. & Benson, M. D. following published PDB structures were used in the paper: 4ODH, 6IC3, 6Z1O, 6HUD, Chemical characterization of a lambda I amyloid protein isolated from 5JZ7, 6QB6, 6Q0E, 7JVA, and 5MUD. The accession codes for the IGLV1-51*02 segment formalin-fixed and paraffin-embedded tissue sections. Amyloid 11,50–55 from the IMGT database are M30446 and from the VBase2 humIGLV015. The IGLJ2 (2004). and IGLJ3 gene segments were taken from GenBank with the Gene IDs 28832 and 28831, 25. Lu, Y., Jiang, Y., Prokaeva, T., Connors, L. H. & Costello, C. E. Oxidative post- respectively. Source data are provided with this paper for the following figures: Fig. 1c, translational modifications of an amyloidogenic immunoglobulin light chain Fig. 5c, d, e, Supplementary Fig. 1a, Supplementary Fig. 2, Supplementary Fig. 5d, and protein. Int. J. Mass Spectrom. 416,71–79 (2017). Supplementary Fig. 6. Source data are provided with this paper. 26. Kumar, S. et al. Assay to rapidly screen for immunoglobulin light chain glycosylation: a potential path to earlier AL diagnosis for a subset of patients. Leukemia 33, 254–257 (2019). Received: 18 June 2021; Accepted: 1 October 2021; 27. Mellors, P. W. et al. MASS-FIX for the detection of monoclonal proteins and light chain N-glycosylation in routine clinical practice: a cross-sectional study of 6315 patients. Blood Cancer J. 11,1–9 (2021). 28. Dispenzieri, A. et al. N-glycosylation of monoclonal light chains on routine MASS-FIX testing is a risk factor for MGUS progression. Leukemia 34, 2749–2753 (2020). References 29. Sagis, L. M., Veerman, C. & van der Linden, E. Mesoscopic properties of 1. Blancas-Mejia, L. M. et al. Immunoglobulin light chain amyloid aggregation. semiflexible amyloid fibrils. Langmuir 20, 924–927 (2004). Chem. Commun. 54.76, 10664–10674 (2018). 30. Adamcik, J. & Mezzenga, R. Study of amyloid fibrils via atomic force 2. Herrera, G. A., Teng, J., Turbat-Herrera, E. A., Zeng, C. & del Pozo-Yauner, L. microscopy. Curr. Opin. Colloid Interface Sci. 17, 369–376 (2012). Understanding Mesangial Pathobiology in AL-amyloidosis and Monoclonal 31. Palladini, G. et al. Circulating amyloidogenic free light chains and serum Immunoglobulin Light Chain Deposition Disease. Kidney Int. Rep. (2020). N-terminal natriuretic peptide type B decrease simultaneously in association 3. Wechalekar, A. D. et al. A European collaborative study of treatment with improvement of survival in AL. Blood 107, 3854–3858 (2006). outcomes in 346 patients with cardiac stage III AL amyloidosis. Blood 121, 32. Poshusta, T. L. et al. Mutations in specific structural regions of 3420–3427 (2013). immunoglobulin light chains are associated with free light chain levels in 4. Merlini, G. et al. Systemic immunoglobulin light chain amyloidosis. Nat. Rev. patients with AL amyloidosis. PloS ONE 4, e5169 (2009). Dis. Prim. 4,1–19 (2018). 33. Thal, D. R., Walter, J., Saido, T. C. & Fändrich, M. Neuropathology and 5. Janeway C. A. Jr., Travers, P., Walport, M. & Shlomchik, M. J. Immunobiology: biochemistry of Aβ and its aggregates in Alzheimer’s disease. Acta the Immune System in Health and Disease. 5th edn (Garland Science, 2001). Neuropathol. 129, 167–182 (2015). The generation of diversity in immunoglobulins. 34. Wulff, M. et al. Enhanced fibril fragmentation of N-terminally truncated and 6. Lefranc, M. P. IMGT®, the international ImMunoGeneTics database. Nucl. pyroglutamyl-modified Aβ peptides. Angew. Chem.Int.Ed. 55,5081–5084 (2016). Acids Res. 29, 207–209 (2001). 35. Stevens, F. J. Four structural risk factors identify most fibril-forming kappa 7. Comenzo, R. L., Zhang, Y., Martinez, C., Osman, K. & Herrera, G. A. The tropism light chains. Amyloid 7, 200–211 (2000). of organ involvement in primary systemic amyloidosis: contributions of Ig VL 36. Omtvedt, L. A. et al. Glycosylation of immunoglobulin light chains associated germ line gene use and clonal plasma cell burden. Blood 98,714–720 (2001). with amyloidosis. Amyloid 7, 227–244 (2000). 8. Perfetti, V. et al. Analysis of Vλ-Jλ expression in plasma cells from primary 37. Solá, R. J. & Griebenow, K. A. I. Effects of glycosylation on the stability of (AL) amyloidosis and normal bone marrow identifies 3r (λIII) as a new protein pharmaceuticals. J. Pharm. Sci. 98, 1223–1245 (2009). amyloid-associated germline gene segment. Blood 100, 948–953 (2002). 10 NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-26553-9 ARTICLE 38. Schwarz, F. & Aebi, M. Mechanisms and principles of N-linked protein 69. Fernandez-Escamilla, A. M., Rousseau, F., Schymkowitz, J. & Serrano, L. glycosylation. Curr. Opin. Struct. Biol. 21, 576–582 (2011). Prediction of sequence-dependent and mutational effects on the aggregation 39. Lee, H. S., Qi, Y. & Im, W. Effects of N-glycosylation on protein conformation of peptides and proteins. Nat. Biotechnol. 22, 1302–1306 (2004). and dynamics: Protein Data Bank analysis and molecular dynamics simulation 70. Maurer-Stroh, S. et al. Exploring the sequence determinants of amyloid study. Sci. Rep. 5,1–7 (2015). structure using position-specific scoring matrices. Nat. Methods 7, 237–242 40. Bellotti, V., Mangione, P. & Merlini, G. Immunoglobulin light chain (2010). amyloidosis—the archetype of structural and pathogenic variability. J. Struct. 71. Garbuzynskiy, S. O., Lobanov, M. Y. & Galzitskaya, O. V. FoldAmyloid: a Biol. 130, 280–289 (2000). method of prediction of amyloidogenic regions from protein sequence. 41. Pepys, M. B. et al. Human serum amyloid P component is an invariant Bioinformatics 26, 326–332 (2010). constituent of amyloid deposits and has a uniquely homogeneous 72. Conchillo-Solé, O. et al. AGGRESCAN: a server for the prediction and glycostructure. Proc. Natl. Acad. Sci. USA 91, 5602–5606 (1994). evaluation of “hot spots” of aggregation in polypeptides. BMC Bioinform. 8,65 42. Liberta, F. et al. Morphological and primary structural consistency of fibrils (2007). from different AA patients (common variant). Amyloid 26, 164–170 (2019). 73. Walsh, I., Seno, F., Tosatto, S. C. & Trovato, A. PASTA 2.0: an improved 43. Schmidt, M. et al. Cryo-EM structure of a transthyretin-derived amyloid fibril server for protein aggregation prediction. Nucl. Ac. Res. 42, W301–W307 from a patient with hereditary ATTR amyloidosis. Nat. Commun. 10,1–9 (2019). (2014). 44. Scheres, S. H., Zhang, W., Falcon, B. & Goedert, M. Cryo-EM structures of tau 74. Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory filaments. Curr. Opin. Struct. Biol. 64,17–25 (2020). research and analysis. J. Comput. Chem. 25, 1605–1612 (2004). 45. Bergström, J. et al. Amyloid deposits in transthyretin derived amyloidosis: cleaved transthyretin is associated with distinct amyloid morphology. J. Acknowledgements Pathol. 206, 224–232 (2005). We thank the Deutsche Forschungsgemeinschaft for funding (projects FA 456/23 and FA 46. Westermark, G. T., Fändrich, M. & Westermark, P. AA amyloidosis: 456/27 to M.F., HA 7138/3 to C.H., HE 8472/1-1 to U.H., HU 2400/1-1 to S.H., SCHO pathogenesis and targeted therapy. Annu. Rev. Pathol. 10, 321–344 (2015). 1364/2-1 to S. O. S., and SCHM 997/30-1 to V.S and a Start-up Grant of the Research 47. Kollmer, M. et al. Cryo-EM structure and polymorphism of Aβ amyloid fibrils Unit FOR 2969 to C.R.). All cryo-EM data were collected at the European Molecular purified from Alzheimer’s brain tissue. Nat. Commun. 10,1–8 (2019). Biology Laboratory, Heidelberg (Germany), funded by iNEXT (Horizon 2020, European 48. Schweighauser, M. et al. Structures of α-synuclein filaments from multiple Union). We are grateful to Karl-Heinz Gührs for critically reading the paper. system atrophy. Nature 585, 464–469 (2020). 49. Bansal, A. et al. AA amyloid fibrils from diseased tissue are structurally different from in vitro formed SAA fibrils. Nat. Commun. 12,1–9 (2021). Author contributions 50. Fändrich, M. & Schmidt, M. Methods to study the structure of misfolded protein L.R., S.K., G.A., J.B., S.W., and C.R. carried out experiments. L.R, S.K. G.A., J.B., M.N., states in systemic amyloidosis. Biochemical Soc. Trans. 49, 977–985 (2021). S.S., N.B., C.H., V.S., S.H., P.W., S.W., C.R., M.S., and M.F. analyzed data. R.M., U.H., 51. Read, C., Walther, P. & von Einem, J. Quantitative Electron Microscopy to and S.O.S. contributed tools, reagents, or materials. M.F. designed research. L.R. and M.F. Study HCMV Morphogenesis, in: Human Cytomegaloviruses: Methods and wrote the paper with contributions from all other authors. Protocols, Methods in Molecular Biology (Ed. Yurochko, A. D.) 265–289 (Springer US, 2021). 52. Read, C., Schauflinger, M., Nikolaenko, D., Walther, P. & von Einem, J. Funding Open Access funding enabled and organized by Projekt DEAL. Regulation of human cytomegalovirus secondary envelopment by a C-terminal tetra-lysine motif in pUL71. J. Virol. 93, e02244-18 (2019). 53. Kremer, J. R., Mastronarde, D. N. & McIntosh, J. R. Computer visualization of Competing interests three-dimensional image data using IMOD. J. Struct. Biol. 116,71–76 (1996). The authors declare no competing interests. 54. Kollmer, M. et al. Electron tomography reveals the fibril structure and lipid interactions in amyloid deposits. Proc. Natl. Acad. Sci. USA 113, 5604–5609 (2016). Additional information 55. Feuz, L., Leermakers, F. A. M., Textor, M. & Borisov, O. Bending rigidity and Supplementary information The online version contains supplementary material induced persistence length of molecular bottle brushes: a self-consistent-field available at https://doi.org/10.1038/s41467-021-26553-9. theory. Macromolecules 38, 8891–8901 (2005). 56. Annamalai, K. et al. Polymorphism of amyloid fibrils in vivo. Angew. Chem. Correspondence and requests for materials should be addressed to Marcus Fändrich. Int. Ed. 55.15, 4822–4825 (2016). Peer review information Nature Communications thanks Gareth J Morgan, Luis del 57. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Pozo-Yauner and the other, anonymous, reviewer(s) for their contribution to the peer Nat. Methods 9, 676–682 (2012). review of this work. 58. Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017). 59. Zhang, K. Gctf: Real-time CTF determination and correction. J. Struct. Biol. Reprints and permission information is available at http://www.nature.com/reprints 193,1–12 (2016). 60. He, S. & Scheres, S. H. W. Helical reconstruction in RELION. J. Struct. Biol. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 198, 163–176 (2017). published maps and institutional affiliations. 61. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of coot. Acta Crystallogr. D. Biol. Crystallogr. 66, 486–501 (2010). 62. Afonine, P. V. et al. Real-space refinement in PHENIX for cryo-EM and Open Access This article is licensed under a Creative Commons crystallography. Acta Crystallogr. D. Struct. Biol. 74, 531–544 (2018). Attribution 4.0 International License, which permits use, sharing, 63. Afonine, P. V. et al. New tools for the analysis and validation of cryo-EM adaptation, distribution and reproduction in any medium or format, as long as you give maps and atomic models. Acta Crystallogr. D. Struct. Biol. 74, 814–840 (2018). appropriate credit to the original author(s) and the source, provide a link to the Creative 64. Williams, C. J. et al. MolProbity: more and better reference data for improved Commons license, and indicate if changes were made. The images or other third party all atom structure validation. Protein Sci. 27.1, 293–315 (2018). material in this article are included in the article’s Creative Commons license, unless 65. Wu, Z. et al. MASH explorer: a universal software environment for top-down indicated otherwise in a credit line to the material. If material is not included in the proteomics. J. Proteome Res. 19, 3867–3876 (2020). article’s Creative Commons license and your intended use is not permitted by statutory 66. Niedermeyer, T. H. & Strohalm, M. mMass as a software tool for the regulation or exceeds the permitted use, you will need to obtain permission directly from annotation of cyclic peptide tandem mass spectra. PloS ONE 7, e44913 (2012). the copyright holder. To view a copy of this license, visit http://creativecommons.org/ 67. Gill, S. C. & Von Hippel, P. H. Calculation of protein extinction coefficients licenses/by/4.0/. from amino acid sequence data. Anal. Biochem. 182, 319–326 (1989). 68. Kabat. E. A., Wu, T. T., Reid-Miller, M., Peny, H. & Gottesman, K. Sequences of Proteins of Immunological Interest. 4th edn (United States Department of © The Author(s) 2021 Health and Human Services, 1987) NATURE COMMUNICATIONS | (2021) 12:6434 | https://doi.org/10.1038/s41467-021-26553-9 | www.nature.com/naturecommunications 11

Journal

Nature CommunicationsSpringer Journals

Published: Nov 5, 2021

There are no references for this article.