Peptides modulating conformational changes in secreted chaperones: from in silico design to preclinical proof of concept

Abstract

Blocking conformational changes in biologically active proteins holds therapeutic promise. Inspired by the susceptibility of viral entry to inhibition by synthetic peptides that block the formation of helix-helix interactions in viral envelope proteins, we developed a computational approach for predicting interacting helices. Using this approach, which combines correlated mutations analysis and Fourier transform, we designed peptides that target gp96 and clusterin, 2 secreted chaperones known to shift between inactive and active conformations. In human blood mononuclear cells, the gp96-derived peptide inhibited the production of TNFalpha, IL-1beta, IL-6, and IL-8 induced by endotoxin by >80%. When injected into mice, the peptide reduced circulating levels of endotoxin-induced TNFalpha, IL-6, and IFNgamma by >50%. The clusterin-derived peptide arrested proliferation of several neoplastic cell lines, and significantly enhanced the cytostatic activity of taxol in vitro and in a xenograft model of lung cancer. Also, the predicted mode of action of the active peptides was experimentally verified. Both peptides bound to their parent proteins, and their biological activity was abolished in the presence of the peptides corresponding to the counterpart helices. These data demonstrate a previously uncharacterized method for rational design of protein antagonists.

Fig. 1.

Known helix–helix interaction and its corresponding periodicity. (A) Residue–residue contact map of 2 antiparallel helices taken from the solved structure of HIV-1 gp41 (PDB: 1ENV) calculated using CSU (60). (B) A typical Fourier transform corresponding to the sum of rows in the 21 by 21 matrix representing the interaction between 2 segments centered around residues 36 and 139 of 1ENV. The peak occurs around periodicity of 3.6 residues.

Fig. 2.

In silico detection of a helix–helix interaction in gp96. (A) Residue–residue contact map for gp96 as predicted by SVMcon (19). (B) Map of Fourier-based scores for the predicted residue–residue contact map of gp96. For a typical Fourier transform that is based on residue–residue contact map predictions, see Fig. S1. Zoomed in views of the contact map predictions and the Fourier based scores are presented in C and D, respectively.

Fig. 3.

Analysis of gp96-II interaction with its parent protein. The gp96 was immobilized directly to a CM5 sensor chip [2,000 resonance units (RU)]. Solution containing 5 different concentrations of gp96-II was injected, and the interaction was monitored for 5 min using surface plasmon resonance. The affinity constant of the interaction between gp96-II and recombinant gp96 was determined by direct kinetic analysis. The 1:1 Langmuir binding model was used to fit kinetic data giving ka = 3.99 × 10³ M⁻¹·s⁻¹ and kd = 8.45 × 10⁻⁴ s⁻¹.

Fig. 4.

Antiinflammatory activity of the gp96-II peptide. (A) PBMC from 3 healthy human donors were stimulated with 50 pg/mL of LPS in the presence of gp96-II (6.6 μM) or the negative control gp96-III (6.6 μM). Results were expressed as percentage with that of LPS-stimulated untreated cells considered as 100%. (B, C, and D) C57Black/6 mice were injected i.p. with the gp96-II peptide (60 μg per mouse) or saline as a vehicle control. One minute later, LPS (10 mg/kg) was injected i.p. Serum was obtained at 90 min for TNFα levels (B). Six hours after the addition of LPS serum was obtained for IL-6 (C) and IFNγ (D). Five mice per group were studied. The mean cytokine levels (pg/mL) ± SEM are shown.

Fig. 5.

Proposed mechanism of action of the active peptides. (A) A conformational change where 2 helices (rectangles) are brought into contact. (B) The conformational change is blocked by adding the active peptide (filled rectangles) that is derived from one of the helices. (C) Preincubation of the active peptide with a peptide corresponding to its counterpart helix (striped rectangles) abolishes the inhibitory effect. (D) Experimental support of the proposed mechanism of action. Preincubation of the gp96-II peptide with an equimolar concentration of gp96-I, a peptide corresponding to its counterpart helix, substantially attenuated its activity.