Interplay between Cα Methylation and Cα Stereochemistry in the Folding Energetics of a Helix-Rich Miniprotein

Abstract

The α-helix is an abundant and functionally important element of protein secondary structure, which has motivated intensive efforts toward chemical strategies to stabilize helical folds. One such method is the incorporation of non-canonical backbone composition through an additional methyl substituent at the C_α atom. Examples of monomers include the achiral 2-aminoisobutyric acid (Aib) with geminal dimethyl substitution and chiral analogues with one methyl and one non-methyl substituent. While Aib and chiral C_α-Me residues are both established helix promoting moieties, their comparative ability in this regard has not been quantitatively investigated. Addressing this gap would help to inform the use of these building blocks in the construction of peptide and protein mimetics as well as provide fundamental insights into consequences of backbone methylation on folding. Here, we report a quantitative comparison of the impacts of Aib and chiral αMe residues on the high-resolution folded structure and folding thermodynamics of a small helical protein. These results reveal a synergistic stabilizing effect arising from the presence of C_α methylation in conjunction with a C_α stereocenter.

Keywords: Heterogeneous backbones; Miniprotein; Peptidomimetic; Proteomimetic; α-Helix.

Figure 1

Sequence and representative model from the NMR structure ensemble of HP35 alongside monomers used in construction of variants. Each residue shown in bold was individually substituted with Ala, Aib, or a chiral αMe residue with side chain matching the replaced residue in the prototype.

Figure 2

Circular dichroism (CD) analysis of HP35 and variants. (A) CD scans at 20 °C. (B) Thermal melts monitored by CD at 222 nm; lines depict fits to a two‐state folding model. All samples consist of 50 μM peptide in 50 mM phosphate at pH 7.0.

Figure 3

(A) Views from the TOCSY spectra of HP35 and variants showing the downfield shift in the resonance frequency of Arg14 H_α accompanying C_α methylation at Lys7. (B, C) Views from a representative model from the NMR structure ensemble of HP35 showing shielding of Arg14 H_α by the side chain of Phe6 (B) and the burial of Lys7 H_α against the loop between helix 1 and 2.

Figure 4

NMR structure ensembles for HP35 and variants Ser15Aib, Ser15αMeSer, Lys30Aib, and Lys30αMeLys. Hydrophobic core side chains are shown as sticks and positions of backbone modification highlighted by spheres. Backbone rmsd (average and standard deviation) for pairwise overlay of each variant ensemble with the ensemble for HP35 is shown in parentheses.

Figure 5

(A) Chemical denaturation of HP35 and indicated variants in the presence of guanidinium chloride (Gnd) monitored by intrinsic Trp fluorescence (excitation at 295 nm; emission monitored at 345 nm and normalized to 398 nm). Samples consisted of 10 μM in 40 mM phosphate at pH 7 and were measured at room temperature. Data points and error bars depict average and standard deviation of 2–3 replicate measurements, and lines show fits to a two‐state folding model. (B) Plot of unfolding free energies (ΔG°_u) for each variant. Uncertainties depicted are standard error from the fit. (C) Scheme for the double mutant cycle thermodynamic analysis with Ala as the reference state.