Exchange, promiscuity, and orthogonality in de novo designed coiled-coil peptide assemblies

Abstract

De novo protein design is delivering new peptide and protein structures at a rapid pace. Many of these synthetic polypeptides form well-defined and hyperthermal-stable structures. Generally, however, less is known about the dynamic properties of the de novo designed structures. Here, we explore one aspect of dynamics in a series of de novo coiled-coil peptide assemblies: namely, peptide exchange within and between different oligomers from dimers through to heptamers. First, we develop a fluorescence-based reporter assay for peptide exchange that is straightforward to implement, and, thus, would be useful to others examining similar systems. We apply this assay to explore both homotypic exchange within single species, and heterotypic exchange between coiled coils of different oligomeric states. For the former, we provide a detailed study for a dimeric coiled coil, CC-Di, finding a half-life for exchange of 4.2 ± 0.3 minutes at a peptide concentration of 200 μM. Interestingly, more broadly when assessing exchange across all of the oligomeric states, we find that some of the designs are faithful and only undergo homotypic strand exchange, whereas others are promiscuous and exchange to form unexpected hetero-oligomers. Finally, we develop two design strategies to improve the orthogonality of the different oligomers: (i) using alternate positioning of salt bridge interactions; and (ii) incorporating non-canonical repeats into the designed sequences. In so doing, we reconcile the promiscuity and deliver a set of faithful homo-oligomeric de novo coiled-coil peptides. Our findings have implications for the application of these and other coiled coils as modules in chemical and synthetic biology.

Fig. 1. Assessing strand exchange of CC-Di using fluorescence-based measurements. (A) One possible scheme of the exchange between a quenched labelled and an unlabelled CC dimer, via a tetrameric steady-state intermediate, to form the fluorescent mixed species. The scheme was created with https://www.biorender.com/. (B) Normalised fluorescence time course plots for the exchange of CC-Di at different concentrations of unlabelled CC-Di. The experiments were carried out at 2 μM of labelled CC-Di. The plots are coloured by the concentration of CC-Di: blue, 20 μM; cyan, 30 μM; green, 50 μM; orange, 100 μM; and red, 200 μM. (C) Plots of the observed pseudo-first-order rate constant (k_obs) for exchange at different concentrations of unlabelled CC-Di (2–200 μM). Data points are shown as the average of 3 independent replicates, error bars are for 1 standard deviation, and the line of best fit is shown in red. (D) Normalised fluorescence time course plots for the exchange of CC-Di at different temperatures (28, 32, 37, and 40 °C). The experiments were carried out at 2 μM of labelled CC-Di and 20 μM of unlabelled CC-Di. The plots are coloured by the temperature: blue, 28 °C; green, 32 °C; orange, 37 °C; and red, 40 °C. (E) Arrhenius plot for the temperature dependence of the rate constants for exchange of CC-Di. Values determined from fits to data shown in Fig. S13–S16 and Table S1. Errors are shown to one standard deviation of independent triplicate measurements. All experiments were carried out at 25 °C in phosphate buffered saline (PBS) at pH 7.4 unless otherwise stated. Observed rate constants (k_obs) were determined by fitting the normalised fluorescence time course profiles to an exponential rise (ESI eqn (2)†).

Fig. 2. Homotypic exchange of CC-Tet. (A) Cartoon for the assumed equilibrium reached when labelled and unlabelled variants of CC-Tet are mixed in a 1 : 10 ratio. For simplicity, excesses of the unlabelled peptide and any intermediate species are omitted. This cartoon was created with https://www.biorender.com/. (B) Cartoon for the anticipated changes in population of the quenched and free FAM-CC-Tet over time where the major species are represented at different time points. (C) Raw data from experiments following the protocol outlined in the text and depicted in panel B. (See ESI for experimental details.) The data were collected for two different samples: (1) a control sample, represented as “C,” which corresponds to 2 μM FAM-CC-Tet in PBS and (2) an exchange sample, represented as “E,” which corresponds to 2 μM FAM-CC-Tet + 20 μM CC-Tet in PBS. The fluorescence measurements were collected at three different time points (1 h, 24 h, and after annealing) and repeated 3 times. All fluorescence data of FAM-CC-Tet mixtures were subsequently normalised against the data collected for the annealed samples where the averaged observed fluorescence of FAM-CC-Tet in buffer was set to zero and the averaged observed fluorescence of 1 : 10 ratio of FAM-CC-Tet to CC-Tet was set to one.

Fig. 3. Summary of exchange across the whole CC basis set. Heat maps for the normalised fluorescence values for all pairs of peptide mixtures. (A) The raw fluorescence data observed in the homotypic exchange of CC-Tet are shown with the values calculated after normalisation in the bar graph on the left. For these data, the annealed control (2 μM FAM-CC-Tet in PBS) intensity was set to zero, and the annealed exchange (2 μM FAM-CC-Tet + 20 μM CC-Tet in PBS) intensity was set to one. These normalised values are the shown in the dummy heatmaps on the right where the labelled CC-Tet position on the y-axis intersects with unlabelled CC-Tet position on the x-axis. (B–D) complete heat maps showing measurements taken at 1 h (B) and 24 h (C) after mixing, and after a subsequent annealing step (D). With a few exceptions, the post-annealed values are much higher than those observed after 1 h and 24 h. The post-annealed data show exchange within the homotypic mixtures (diagonal values), between CC-Tri and CC-Tet, and between type II coiled coils (CC-Pent 2, CC-Hex 2, and CC-Hept).

Fig. 4. Improved orthogonality between CC-Tri and CC-Tet*. (A) Helical-wheel representations of the heptad-repeat sequences of CC-Tet and CC-Tet*. In CC-Tet, residues that promote interhelical salt-bridge interactions (lysine and glutamic acid) are positioned at the e and g positions, whereas in CC-Tet* these are at b and c. (B) Slices through the X-ray crystal structures of CC-Tet and CC-Tet* show the structural consequences of the different placements of lysine and glutamic acid residues. Except for these, side chains are omitted for clarity. (C) Normalised fluorescence date for exchange between CC-Tri and CC-Tet*. After annealing, values for the hetero-mixtures (off diagonal) are lower than those for the homomers, indicating orthogonality over the CC-Tri/CC-Tet combination (Fig. 3C).

Fig. 5. Mixing heptad and hendecad repeats to improve orthogonality. (A–C) Fluorescence-based orthogonality screens of α-helical-barrel variants incorporating hendecad repeats, CC-Pent2-hen2, CC-Hex2-hen2, and CC-Hept-hen2, where the h positions were Ala, Leu, or Ile. This initial screen was performed against FAM-labelled variants of the parent, heptad-based peptides. Compared with homotypic controls (i.e. FAM-CC-Pent2 + CC-Pent2, FAM-CC-Hept + CC-Hept), the CC-Pent2-hen2 and CC-Hept-hen2 variants showed marked decreases in fluorescence indicating less exchange and thus improved orthogonality (D and F). However, the CC-Hex2 variants still showed some cross-exchange and promiscuity (E). Note: some of the hen2 variants were not stable up to 95 °C and precipitated from solution after annealing as indicated by the striped data columns. (D–F) An AlphaFold2 (ref. 68) model and X-ray crystal structures of CC-Pent2-hen2, CC-Hex2-hen2, and CC-Hept-hen2 (all with Ala at h) respectively are shown with the a positions coloured red, d in green, and h in lilac.

Fig. 6. An Orthogonal CC basis set. (A) X-ray crystal structures and an AlphaFold2 (ref. 68) model for the assembled peptides in this set. The pentamer and heptamer were designed in this study, the other peptides have been published elsewhere. The heat map shown in panel B represents the ideal case where all peptides are orthogonal to one another, i.e. only homotypic exchange is observed. Heat maps from the post-annealing fluorescence exchange data for the CC basis set (C) and the Orthogonal CC basis set (D) are shown side-by-side for comparison. The values in panel D indicate considerable orthogonality across the new set: the homotypic exchange (diagonal) values are all higher than those for any of the heterotypic exchange experiments (off-diagonal). All mixtures containing CC-Hept-IV-hen2 and/or FAM-CCHept-IV-hen2 were annealed to 75 °C instead of 95 °C as these peptides were not stable (with respect to precipitation) at the higher temperature (see Fig. S47†).