Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Aug 27;121(35):e2408554121.
doi: 10.1073/pnas.2408554121. Epub 2024 Aug 22.

Atomic resolution map of the solvent interactions driving SOD1 unfolding in CAPRIN1 condensates

Rashik Ahmed  1   2   3   4 , Mingyang Liang  3 , Rhea P Hudson  4 , Atul K Rangadurai  1   2   3   4 , Shuya Kate Huang  1   2   3   4 , Julie D Forman-Kay  3   4 , Lewis E Kay  1   2   3   4
Affiliations

Atomic resolution map of the solvent interactions driving SOD1 unfolding in CAPRIN1 condensates

Rashik Ahmed et al. Proc Natl Acad Sci U S A. .

Abstract

Biomolecules can be sequestered into membrane-less compartments, referred to as biomolecular condensates. Experimental and computational methods have helped define the physical-chemical properties of condensates. Less is known about how the high macromolecule concentrations in condensed phases contribute "solvent" interactions that can remodel the free-energy landscape of other condensate-resident proteins, altering thermally accessible conformations and, in turn, modulating function. Here, we use solution NMR spectroscopy to obtain atomic resolution insights into the interactions between the immature form of superoxide dismutase 1 (SOD1), which can mislocalize and aggregate in stress granules, and the RNA-binding protein CAPRIN1, a component of stress granules. NMR studies of CAPRIN1:SOD1 interactions, focused on both unfolded and folded SOD1 states in mixed phase and demixed CAPRIN1-based condensates, establish that CAPRIN1 shifts the SOD1 folding equilibrium toward the unfolded state through preferential interactions with the unfolded ensemble, with little change to the structure of the folded conformation. Key contacts between CAPRIN1 and the H80-H120 region of unfolded SOD1 are identified, as well as SOD1 interaction sites near both the arginine-rich and aromatic-rich regions of CAPRIN1. Unfolding of immature SOD1 in the CAPRIN1 condensed phase is shown to be coupled to aggregation, while a more stable zinc-bound, dimeric form of SOD1 is less susceptible to unfolding when solvated by CAPRIN1. Our work underscores the impact of the condensate solvent environment on the conformational states of resident proteins and supports the hypothesis that ALS mutations that decrease metal binding or dimerization function as drivers of aggregation in condensates.

Keywords: client/scaffold; folding-unfolding equilibria; methyl-TROSY NMR; phase separation; protein stability.

PubMed Disclaimer

Conflict of interest statement

Competing interests statement:The authors declare no competing interest.

Figures

Fig. 1.
Fig. 1.
Stress granule “scaffold” protein CAPRIN1 solvates the SOD1 “client” and biases it toward the unfolded ensemble. (A) Schematic depicting the solvation of a “client” protein in a “scaffold” protein condensate. Tyrosine residues are highlighted on CAPRIN1 as these play important roles as stickers in phase separation (15, 33), and methyl groups are shown on SOD1, as these serve as structural probes in the present work. (B) Maturation pathway of SOD1. In the absence of metals and disulfide bond formation involving C146 and C57, E, E S-H SOD1 is in equilibrium between an unfolded ensemble and a folded conformation consisting of an 8-stranded β-barrel with two long loops. Upon Cu, Zn metalation and intrasubunit disulfide bond formation, Cu, Zn S-S SOD1 adopts a homodimeric state. The pseudo-WT (pWT) construct used in this study, with cysteine mutants C6A and C111S color-coded in black on the SOD1 structure, is shown (34) (PBD ID: 1HL5). The electrostatic loop and the zinc (dimer) loop are shown in blue and pink, respectively, the intrasubunit disulfide bridge is shown in green, and the zinc and copper metals are depicted as purple and orange spheres, respectively. (C) Schematic of the construct and associated mutations/truncation of CAPRIN1 used in this study. (D) Selected regions of [1H, 13C] HMQC spectra of the methyl region of 300 µM E, E S-H pWT SOD1 in the absence (purple) and presence (blue) of 3.5 mM CAPRIN1, 25 °C, 800 MHz. 1D slices of representative peaks are shown, highlighting the shift in population of folded and unfolded conformers upon addition of CAPRIN1. (E) Population of SOD1 folded conformation as a function of CAPRIN1 concentration, 25 °C. Population fractions were calculated from peak volumes in [1H,13C] HMQC spectra, taking into account differences in transverse relaxation rates of magnetization derived from folded and unfolded species (Eq. [S3] in SI Appendix, Materials and Methods). (F) Population of SOD1 folded conformation in the presence of 43.3 mg/mL of various crowders, based on the integrals of 19F resonances of fluorotryptophan-labeled SOD1 derived from folded and unfolded conformers shown in SI Appendix, Fig. S1 A, D–G. Error bars report SD of replicates (circles). Crowders are annotated based on their net charge at pH 7, either neutral, positive (+), or negative (−). The molar concentrations of the crowders are CAPRIN1 (4.0 mM), Lysozyme (3.0 mM), BSA (0.63 mM), and ficoll 70 (0.62 mM).
Fig. 2.
Fig. 2.
Intermolecular interactions of CAPRIN1 with folded and unfolded E, E S-H pWT SOD1. (A) Cartoon depicting the conjugation of a maleimide DOTA cage onto an exogenously introduced CAPRIN1 cysteine. Cages are filled with either paramagnetic (gadolinium) or diamagnetic (lutetium) metal. (B) Overlay of selected slices through peaks derived from unfolded (green) and folded (purple) E, E S-H pWT SOD1 for two representative residues, I99 and G114, in diamagnetic (darker colors, dashed lines) and paramagnetic (lighter colors, solid lines; spin-label at position 615) spectra. (C) Intensity profiles, Ipara/Idia, for 300 µM 15N E, E S-H pWT SOD1 in the presence of 100 µM 14N CAPRIN1 S615C-maleimide-DOTA coordinated with either gadolinium (Ipara) or lutetium (Idia). Error bars report SD of replicates (SI Appendix, Materials and Methods). Intermolecular Ipara/Idia ratios measured for folded (unfolded) E, E S-H pWT SOD1 are shown in purple (green). Only residues with unique folded and unfolded resonances are shown. The solid gray line indicates the charge of SOD1 averaged over nine residues (±4 residues on either side) with the dashed gray line indicating a charge of 0. A schematic diagram of CAPRIN1, with arginine-rich (green rectangles) and aromatic-rich (purple rectangles) regions highlighted along with the cysteine mutation for the maleimide-DOTA cage conjugation site (blue circle), and the secondary structure elements of folded SOD1, are displayed above the intermolecular PRE profile. Electrostatic loop is abbreviated as (E)-loop. (D) Linear correlation plot of Ipara/Idia ratios for folded and unfolded E, E S-H pWT SOD1. The dashed line is y = x. (E)–(H) As (C) and (D), except with 100 µM 14N CAPRIN1 A658C-maleimide-DOTA (E) and (F) or 100 µM 14N CAPRIN1 S678C-maleimide-DOTA (G) and (H). (I) Intermolecular Ipara/Idia profiles for 250 µM 15N E, E S-H pWT SOD1 in the presence of 250 µM 14N CAPRIN1 A658C-maleimide-DOTA (purple bars). Values derived from all nonoverlapping folded SOD1 resonances and resonances from loop regions with identical folded and unfolded chemical shifts are shown. The recognition factor, an aggregate measure of the hydrophilicity and average interaction energy of an amino acid (45), is computed for the SOD1 amino acid sequence, normalized, averaged over a 9-residue window and plotted as 1 – Normalized Recognition Factor (black circles). (J) Heat map of CAPRIN1 interaction sites in folded SOD1 plotted onto the structure of a single protomer from the mature SOD1 structure (PDB ID: 1HL5) (34). The relative strengths of the intermolecular interactions, used for the heat map, were calculated as 1-IparaIdia1-IparaIdiamaxx100%, based on the Ipara/Idia profile shown in (I). All NMR experiments were recorded at 25 °C, 1 GHz.
Fig. 3.
Fig. 3.
Effect of the CAPRIN1 condensed phase on the folding equilibrium and dynamics of E, E S-H pWT SOD1. (A)–(B) [1H, 15N] HSQC (A) and [1H, 13C] ddHMQC (B) spectra of 2H, 15N, 13C-ILV E, E S-H pWT SOD1 in dilute (blue) and condensed (orange) phases, 25 °C, 800 MHz. Vertical traces in A (left of 2D spectra) at the position of the dashed line illustrate that [1H, 15N] correlations from the folded state are observed in the dilute phase, but not in the condensed phase. In contrast, expected correlations for the folded conformer are observed in the ddHMQC spectra. (C) Quantification of folded and unfolded E, E S-H pWT SOD1 fractional populations based on integrals of resonances from folded and unfolded conformers recorded under fully relaxed conditions and taking into account the enhanced transverse relaxation rates of correlations in the condensed phase based on rates listed in panel F (SI Appendix, Materials and Methods). (D) Concentrations of folded and unfolded states of E, E S-H pWT SOD1 in the dilute (blue) and condensed (orange) phases from comparison to a reference spectrum with a known total concentration of protein. (E) Per-residue 15N R2 rates for the unfolded E, E S-H pWT SOD1 ensemble or resonances within disordered regions of the folded conformer in the dilute (blue) and condensed (orange) phases. Secondary structure elements of SOD1 are indicated above. (F) [1H-13C] multiple quantum relaxation rates of Ile δ1 methyl groups in folded E, E S-H pWT SOD1 in dilute (blue) and condensed (orange) phases. The average R2 increase reported pertains only to the folded isoleucine resonances. A similar ratio is reported for Ile δ1 methyl groups from the unfolded ensemble (Right), however, only an average is obtained as the resolution in the current spectrum is not sufficient to resolve the correlations.
Fig. 4.
Fig. 4.
Coupled unfolding of SOD1 and aggregation of E, E S-H pWT SOD1:CAPRIN1 condensed phase. (A) A CAPRIN1:E, E S-H pWT SOD1 phase-separated sample at T = 0 d (left) and at T = 62 d (right), showing the formation of aggregates at the longer time point. The sample was stored at 4 °C when not in use for NMR measurements. (B) 1H NMR spectra of the sample at T = 0 d (black) and at T = 62 d (orange), indicating loss of soluble NMR-observable CAPRIN1 (gray shaded region; note that the variant of CAPRIN1 used does not have ILV residues) and loss of E, E S-H pWT SOD1 (pink shaded region). (C) 1H NMR spectrum of CAPRIN1 condensed phase (no SOD1) at T = 0 (black) and at T = 20 mo (orange dashed line), indicating little change after storage at 4 °C. (D) 1H NMR spectrum of 200 µM 2H, 13CH3-ILV SOD1 in NMR buffer at T = 0 (black) and at T = 86 d after storage at 4 °C (orange dashed line). (E) Relative integrals of NMR-observable signals at T = 62 d vs. T = 0 d for CAPRIN1 in a CAPRIN1 condensed phase, E, E S-H pWT SOD1 in buffer, and CAPRIN1 and E, E S-H pWT SOD1 in a CAPRIN1:E, E S-H pWT condensed phase (left to right). The rate of change of the NMR observable signal is assumed to be linear between T = 0 and the measured time point in the CAPRIN1 condensed phase and E, E S-H pWT SOD1 in buffer for computing the expected signal at T = 62 d. (F)–(H) Negative-stain TEM images of (F) the CAPRIN1:E, E S-H pWT SOD1 condensed phase, (G) the CAPRIN1:E, E S-H pWT SOD1 dilute phase and (H) the CAPRIN1 condensed phase. (All scale bars are 200 nm.)
Fig. 5.
Fig. 5.
Metalated SOD1 is more resilient to unfolding when solvated by CAPRIN1 relative to E, E S-H SOD1. (A) Schematic indicating the conformational states of SOD1 produced in the absence of zinc or CAPRIN1 (purple) and in the presence of excess zinc (green). In the absence of zinc, SOD1 exists in equilibrium between folded and unfolded monomeric states (E, E S-H SOD1monomer). In the presence of excess zinc, SOD1 is primarily a folded dimer with full zinc occupancy of the primary, high-affinity, zinc-binding site, and high zinc occupancy at the secondary, low-affinity metal binding site (Zn, Zn S-H SOD1Dimer, ≈ 75%; see SI Appendix). Other minor states are shown in the diagram and are in equilibrium with the major zinc form. (B) Selected region of [1H,13C] ddHMQC spectra, 800 MHz, 25 °C, of 333 µM pWT SOD1 in the presence of 5 mM ZnCl2 (green), without (left) and with (right) 3.5 mM CAPRIN1. The corresponding ddHMQC spectra of pWT SOD1 (300 µM; purple dashed) are reproduced here from Fig. 1D for comparison. In the presence of CAPRIN1 (right, green), the populations of the minor states are elevated, giving rise to resonances that overlap with those from the metal-free pWT SOD1 state (purple dashed). (C) Population of pWT SOD1 folded conformations, including minor states, as a function of CAPRIN1 concentration. The metal-free pWT SOD1 vs. CAPRIN1 profile (purple) is reproduced from Fig. 1E for comparison.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Shin Y., Brangwynne C. P., Liquid phase condensation in cell physiology and disease. Science 357, eaaf4382 (2017). - PubMed
    1. Banani S. F., Lee H. O., Hyman A. A., Rosen M. K., Biomolecular condensates: Organizers of cellular biochemistry. Nat. Rev. Mol. Cell Biol. 18, 285–298 (2017). - PMC - PubMed
    1. Li P., et al. , Phase transitions in the assembly of multivalent signalling proteins. Nature 483, 336–340 (2012). - PMC - PubMed
    1. Zhang Y., Narlikar G. J., Kutateladze T. G., Enzymatic reactions inside biological condensates. J. Mol. Biol. 433, 166624 (2021). - PMC - PubMed
    1. Su X., et al. , Phase separation of signaling molecules promotes T cell receptor signal transduction. Science 352, 595–599 (2016). - PMC - PubMed

LinkOut - more resources