Secondary Structure in the Core of Amyloid Fibrils Formed from Human β2m and its Truncated Variant ΔN6

Amyloid fibrils formed from initially soluble proteins with diverse sequences are associated with an array of human diseases. In the human disorder, dialysis-related amyloidosis (DRA), fibrils contain two major constituents, full-length human β2-microglobulin (hβ2m) and a truncation variant, ΔN6 which lacks the N-terminal six amino acids. These fibrils are assembled from initially natively folded proteins with an all antiparallel β-stranded structure. Here, backbone conformations of wild-type hβ2m and ΔN6 in their amyloid forms have been determined using a combination of dilute isotopic labeling strategies and multidimensional magic angle spinning (MAS) NMR techniques at high magnetic fields, providing valuable structural information at the atomic-level about the fibril architecture. The secondary structures of both fibril types, determined by the assignment of ∼80% of the backbone resonances of these 100- and 94-residue proteins, respectively, reveal substantial backbone rearrangement compared with the location of β-strands in their native immunoglobulin folds. The identification of seven β-strands in hβ2m fibrils indicates that approximately 70 residues are in a β-strand conformation in the fibril core. By contrast, nine β-strands comprise the fibrils formed from ΔN6, indicating a more extensive core. The precise location and length of β-strands in the two fibril forms also differ. The results indicate fibrils of ΔN6 and hβ2m have an extensive core architecture involving the majority of residues in the polypeptide sequence. The common elements of the backbone structure of the two proteins likely facilitates their ability to copolymerize during amyloid fibril assembly.

optimized phase and pulse length was applied during indirect evolution and acquisition periods for all experiments 1 . Chemical shifts were calibrated relative to DSS, using adamantine as a secondary standard. The temperature of approximately 275 K for all experiments was maintained by a stream of nitrogen gas from either the Kinetics Thermal System XR air-jet system (Stone Ridge, NY) on the 750 MHz spectrometer or the Bruker BCU system (Bruker BioSpin, Billerica, MA) on the 800 MHz and 900 MHz spectrometers.
For 1D experiments, a 13 C-1 H CP contact time of ~ 1.5 ms was used. For DP and INEPT, a recycle delay of 5-5.5 s was used to allow sufficient relaxation.
For 2D 13 C-13 C experiments, broadband RFDR spectra were acquired at 18 kHz or 20 kHz spinning frequency. A 6 µs π pulse was placed in the middle of each rotor cycle of the rotorsynchronized RFDR mixing period to recouple the 13 C-13 C dipolar coupling. A 100 kHz continuouswave (CW) decoupling was applied to the 1 H channel during the RFDR period. A total mixing time, τ RFDR , of 1.6 ms was used to establish adjacent 13 C-13 C correlations. In the case of band-selective RFDR, a relatively weak 12.5 µs π pulse was used to excite the aliphatic resonances and a long τ RFDR of 16.2 ms was used for the neighboring Cα-Cα correlations. A 32-step phase cycling was used to compensate the chemical shift offsets and rf inhomogeneity for weak π pulses.
For 2D 15 N-13 C experiments, the ZF TEDOR is a modified TEDOR experiments consisting of two z-filter periods, removing unwanted multiple-quantum and anti-phase spin coherences from spin evolution 13 C-13 C J couplings. All ZF TEDOR experiments were conducted at ω r /2π of 12.5 kHz and 18 kHz spinning frequencies on 750 MHz and 800 MHz spectrometers ( 1 H frequency), respectively. Two π S3 pulses (a pulse length of 12 µs for each one) per rotor period were applied on the 15 N channel and an xy-4 phase cycling scheme was used. The total dipolar recoupling time (τ TEDOR ) of 1.6 ms and 6.4 ms was used for one-bond and multi-bond 15 N-13 C correlations. 1  Each experiment has 8 transients and two consecutive experiments were conducted and added for better signal average, giving a total of ~ 7-10 days measurement time for each 3D spectrum.
In contrast, simultaneous 15 N-13 CO and 15 N-13 Cα transfers in TEDOR-CC were achieved using ZF TEDOR 2,3 . This pulse program was designed as described in our recent work and was successfully applied for the assignment of drug-resistant S31N M2 proton transporter from influenza A 4 . After onebond 15 N-13 C transfer, 13 C-13 C magnetization transfer was achieved through a RFDR recoupling period of 4.8 ms in the current study. Phase alternations of xy-4 and xy-16 were used for TEDOR mixing and RFDR pulses, respectively. We measured TEDOR-CC spectra under 20 kHz MAS on the 900 MHz spectrometer. Optimized linear field compensation was applied to correct the field drift. The 15 N spectral width was 6.7 kHz and the maximum t 1 evolution time was 11.1 ms, corresponding to 148 t 1 points. The S4 13 C spectral width was 20 kHz and the number of t 2 points was 354, giving a maximum t 2 of 8.9 ms. An 83 kHz 1 H decoupling field was used during TEDOR mixing and detection. A 100 kHz CW 1 H decoupling was used in the period of the 6-ms π pulse in RFDR mixing. The total experiment time was approximately 5 days. NCOCX and NCACX regions were extracted respectively from the single TEDOR-CC spectrum.

S5
S6 SI Table 1. NMR chemical shifts of P32 and C80 of hβ 2 m and ∆N6 in native and fibril forms. Chemical shift of native hβ 2 m and ∆N6 proteins were taken from 5 . C-13 C planes at 15 N of P32 from ∆N6 fibrils (133.4 ppm) are shown. All experiments utilize 13 C-1 H CP to generate the initial magnetization, which significantly enhance the intensity of proline. NCOCX and NCACX were obtained using the TEDOR-CC experiment, which utilizes TEDOR scheme for N-C magnetization transfer. The obtained chemical shift identified its trans-conformation. Figure 4. Identification of bond conformation of H31-P32 in hβ 2 m and ∆N6 fibrils by comparing the C'/Cβ/Cγ chemical shift to folded proteins from the biological magnetic resonance bank (BMRB). Histograms show secondary chemical shift of C' and Cβ and chemical shift of Cγ for Pro residues preceded by a cis (red) or trans (blue) bond conformation. The histogram is reproduced with permission from Shen and Bax 6 . Secondary chemical shifts of C' and Cβ are calculated relative to the random coil values used in TALOS+ program. As indicated by violet and green arrows, respectively, for hβ 2 m and ∆N6, H31-P32 adopts a trans-conformation for both fibril types. Figure 5. Representative 2D and 3D spectra of hβ 2 m fibrils to show the assignment of C80. (a) PAIN-CP spectra of [U-15 N-and 1,3-13 C 2 -glycerol]-labeled hβ 2 m. Violet solid lines guide the sequential connectivity of Y78-A79-C80-R81-V82. Pink stripes show the connectivity between S52 with its neighboring residues, E50, H51, D53 and L54, indicating the efficient correlation of the PAIN-CP experiment. (b) Backbone walks from A79 to V82. 2D and 3D spectra were acquired on 900 and 750 MHz spectrometers, respectively. Figure 6. Identification of redox state of C80 in fibrils formed from hβ 2 m and ∆N6 by analysis of their Cβ chemical shifts. Histograms show the distribution of Cβ chemical shifts of oxidized (red) and reduced (blue) cysteines in different proteins. The histogram is reproduced with permission from Sharma and Rajarathnam 7 . As indicated by violet and green arrows for hβ 2 m and ∆N6 fibrils, respectively, the chemical shifts of Cβ for C80 are consistent with an oxidized state, suggesting that the disulfide bond between C25 and C80 is intact in both fibril types.