Zinc binding regulates amyloid-like aggregation of GAPR-1

Abstract

Members of the CAP superfamily (Cysteine-rich secretory proteins, Antigen 5, and Pathogenesis-related 1 proteins) are characterized by the presence of a CAP domain that is defined by four sequence motifs and a highly conserved tertiary structure. A common structure-function relationship for this domain is hitherto unknown. A characteristic of several CAP proteins is their formation of amyloid-like structures in the presence of lipids. Here we investigate the structural modulation of Golgi-Associated plant Pathogenesis Related protein 1 (GAPR-1) by known interactors of the CAP domain, preceding amyloid-like aggregation. Using isothermal titration calorimetry (ITC), we demonstrate that GAPR-1 binds zinc ions. Zn²⁺ binding causes a slight but significant conformational change as revealed by CD, tryptophan fluorescence, and trypsin digestion. The Zn²⁺-induced conformational change was required for the formation of GAPR-1 oligomers and amyloid-like assemblies in the presence of heparin, as shown by ThT fluorescence and TEM. Molecular dynamics simulations show binding of Zn²⁺ to His⁵⁴ and His¹⁰³ Mutation of these two highly conserved residues resulted in strongly diminished amyloid-like aggregation. Finally, we show that proteins from the cysteine-rich secretory protein (CRISP) subfamily are also able to form ThT-positive structures in vitro in a heparin- and Zn²⁺-dependent manner, suggesting that oligomerization regulated by metal ions could be a common structural property of the CAP domain.

Keywords: Amyloid; CAP superfamily; CRISP; GAPR-1; Heparin; Zinc.

Figure 1. Zn²⁺ binding to GAPR-1

(A) ITC thermogram of Zn²⁺ titration into 60 μM GAPR-1. The upper panel represents the corrected heat rate. The lower panel shows the released heat compared with molar ratio of Zn²⁺ to GAPR-1. (B) ITC thermogram of Ca²⁺ titration into GAPR-1. (C) Far-UV CD spectra of 15 μM GAPR-1, recorded in the absence (solid line) and presence (dashed line) of 100 μM Zn²⁺. (D) Intrinsic tryptophan fluorescence of GAPR-1 in the absence or presence of Zn²⁺ (0–500 μM, as indicated). EDTA (1 mM) was subsequently added to the sample containing 500 μM Zn²⁺. Fluorescence intensities were normalized to the initial fluorescence intensity of GAPR-1 in the absence of Zn²⁺. The results represent the means (± S.D.) of three independent experiments. Statistical significance was determined by Student’s t-test; **P < 0.01.

Figure 2. Zn²⁺ binding induces a conformational change in GAPR-1

(A) Western blot analysis of limited trypsin digestion of GAPR-1 using N-terminal GAPR-1 antibody. GAPR-1 (30 μM) was incubated in the presence of Zn²⁺ (0–500 μM) at 37°C for 30 min, followed by incubation without (upper panel) or with (lower panel) trypsin in the molar ratio of 1:50 (trypsin: GAPR-1) for 30 min at 37°C. (B) Western blot analysis of limited trypsin digestion of GAPR-1 incubated with 100 μM Zn²⁺ using five different GAPR-1 antibodies (see Table 2, ‘Materials and methods’ section).

Figure 3. Zn²⁺ induces heparin-mediated GAPR-1 amyloid-like aggregation

(A) Kinetics of ThT fluorescence enhancement of 15 μM GAPR-1 incubated with 37.5 μM heparin and 100 μM Zn²⁺. GAPR-1 incubated in the absence of heparin and/or Zn²⁺ and in the presence of 1 mM EDTA are shown as controls. (B) ThT fluorescence enhancement of 15 μM GAPR-1 incubated with 37.5 μM heparin in the presence of increasing concentrations of Zn²⁺ (0–1000 μM).

Figure 4. GAPR-1 forms oligomeric and amyloid-like structures in a zinc- and heparin-dependent manner

(A) Western blot analysis of GAPR-1 (15 μM) after incubation in the absence or presence of 37.5 μM heparin and/or 100 μM Zn²⁺ at 37°C for 0, 2, 7, and 20 h, respectively. (B) Transmission electron micrographs of 15 μM GAPR-1 following incubation with 37.5 μM heparin and 100 μM Zn²⁺ at 37°C for 6 h (a), 18 h (b), and 3 days (c). GAPR-1 incubated with heparin in the absence of Zn²⁺ for 18 h (d). Scale bars represent 50 nm in panels (a,b,d), and 500 nm in panel (c).

Figure 5. Co-ordination of zinc to GAPR-1 by His54 and His103

(A) Zn²⁺ binding site was predicted via ZincBinder. Zn²⁺ binding favors a tetrahedral geometry hierarchy and each residue in the potential binding site was analyzed for binding affinity. (B) Comparison of the 3D structure of GAPR-1 with the predicted Zn²⁺-binding site to the structure of Zn²⁺-bound GLIPR1. The RMSD between the crystals of Zn²⁺-bound GLIPR1 and GAPR-1 was less than 1.937 Å, indicative of a high similarity. (C) Comparison of the distance between the two histidine residues and Zn²⁺ in GLIPR1 and in GAPR-1. Left panel represents the tetrahedral geometry of Zn²⁺ binding in GLIPR1 and GAPR-1; right panel illustrates the superposition of the two Zn²⁺ binding residues in GLIPR1 and GAPR-1 and shows a rotation by 1.2 Å for His⁵⁴ and 1.6 Å for His¹⁰³ in GAPR-1 relative to His⁷⁹ and His¹³⁷ in GLIPR1, respectively. Each residue was labeled according to the crystallographic numbering. W represents a water molecule. (D) Physiochemical properties of zinc binding between GLIPR1 and GAPR-1 were compared. K_DEEP was utilized for analyzing the dissociation constant (pKd), the binding energy (ΔΔG), and the binding efficiency (LE). The corresponding score for GLIPR1 and GAPR-1 was plotted as a bar graph.

Figure 6. Mutation of the putative GAPR-1 metal-binding pocket affects amyloid formation

(A) Kinetics of ThT fluorescence enhancement of 15 μM GAPR-1 WT, H54A or H103A, respectively, incubated with 5 μM heparin and 100 μM Zn²⁺ at 37°C. (B) Western blot analysis of limited trypsin digestion of GAPR-1 WT, H54A and H103A incubated with 100 μM Zn²⁺, using N-terminal GAPR-1 antibody.

Figure 7. Zn²⁺ induces mouse CRISP2 and mouse CRISP4 amyloid formation

A total of 2.5 μM GAPR-1, mouse CRISP2, human CRISP3 or mouse CRISP4 was incubated with 6.25 μM heparin in absence and presence of 100 μM Zn²⁺ at 37°C for 20 h. ThT fluorescence assay was performed and the ThT fluorescence was expressed as the difference between the end point and the starting point (Δ fluorescence). The results represent the means (± S.D.) of three independent experiments.