Mutually stabilizing interactions between proto-peptides and RNA

Abstract

The close synergy between peptides and nucleic acids in current biology is suggestive of a functional co-evolution between the two polymers. Here we show that cationic proto-peptides (depsipeptides and polyesters), either produced as mixtures from plausibly prebiotic dry-down reactions or synthetically prepared in pure form, can engage in direct interactions with RNA resulting in mutual stabilization. Cationic proto-peptides significantly increase the thermal stability of folded RNA structures. In turn, RNA increases the lifetime of a depsipeptide by >30-fold. Proto-peptides containing the proteinaceous amino acids Lys, Arg, or His adjacent to backbone ester bonds generally promote RNA duplex thermal stability to a greater magnitude than do analogous sequences containing non-proteinaceous residues. Our findings support a model in which tightly-intertwined biological dependencies of RNA and protein reflect a long co-evolutionary history that began with rudimentary, mutually-stabilizing interactions at early stages of polypeptide and nucleic acid co-existence.

Fig. 1. Cationic depsipeptides and polyesters are generated in dry-down reactions.

Depsipeptide and polyester mixtures used in this study were generated via dry-down reactions, in the absence of condensing agents, from binary mixtures of a α-amino acids and b α-hydroxy acids. Cationic side chain moieties are blue. c Scheme showing some potential products of a dry-down reaction of the cationic α-hydroxy acid monomer hah. The major species observed by ¹H-NMR were acylated at the α-hydroxy position and free at the ε-amine position, corresponding to linear oligomers having a backbone topology similar to biological proteins, but with ester bonds in place of amide bonds. d ¹H-NMR spectrum of the product mixture resulting from a hah dry-down at 85 °C for 7 days. Integration of the free α-proton indicated that 56% of hah was incorporated into oligomers. The downfield ε-protons at ~3.1 ppm (12% by integration) likely correspond to ε-amidation, in analogy to chemical shift patterns observed upon ε-amidation of Lys in dry-down reactions. Some resonances corresponding to acylated α-species are obscured by the water peak and are not shown, but can be observed by COSY analysis (Supplementary Fig. 8). e Positive-mode ESI-MS spectra showing the production of cationic polyesters via dry-down reaction of hah at 85 °C for 7 days. Labeled species correspond to [M + H]⁺ ions.

Fig. 2. Cationic proto-peptide mixtures generated in dry-down reactions stabilize duplex RNA.

Dry-down reaction products were dissolved in deionized water to give a 100 mM stock solution based on the amount of monomers used at the start of the dry-down. The mixtures contain diverse oligomers at varying abundances, so the concentration of a given oligomer in the mixture would be substantially lower. Control dry-down reactions contained the amino acid alone, with no hydroxy acid (AA control). a Schematic depicting the process used here to generate proto-peptide mixtures. RNA stability studies used either crude dry-down product mixtures, or dialyzed mixtures from which unreacted monomers and short oligomers had been removed with a 500–1000 Da cut-off membrane. For panels b–d, the experiments employed a 10-mer RNA duplex (5′-6-FAM-rCrGrCrUrArArArUrCrG-3′ & 5′-rCrGrArUrUrUrArGrCrG-3IABkFQ-3′, 2.5 μM strand) in buffer (100 mM MES-TEA, 2.5 mM NaCl, pH 7.5). The final pH of the samples was between 5.6 and 6.8, unless otherwise noted. b Thermal denaturation curves for the RNA duplex in the presence of various Arg-containing dry-down reaction mixtures or control mixtures. c Changes in RNA duplex T_m relative to the corresponding amino acid control (dry-down reaction of the amino acid without a hydroxy acid) for crude dry-down mixtures. d Changes in RNA duplex T_m upon addition of dialyzed depsipeptide oligomers. Data are shown as a scatter plot of duplicate experiments. The non-cationic glc + Ala oligomers were included as a control. For the RNA alone condition, three technical replicates from each of the duplicate experiments are shown, for a total of six data points. e Changes in RNA duplex T_m values upon addition of crude polyester and depsipeptide mixtures obtained by drying hab or hah, either alone or with Gly or Ala (at a 1:1 molar ratio), or non-dried controls. Data are shown as a scatter plot of two independent experiments. f Structures of polyesters derived from drying hab or hah, showing potential routes of oligomer degradation via intramolecular O,N acyl transfer.

Fig. 3. Structure–function studies of cationic depsipeptides in stabilizing an RNA duplex.

Each sample contained RNA duplex 1 (5′-rCrGrCrUrArArArUrCrG-3′ and 5′-rCrGrArUrUrUrArGrCrG-3′, 2.5 μM strand) and peptide/depsipeptide (100 μM) in buffered solution (10 mM phosphate, 100 mM NaCl, pH 7.0 or 10 mM acetate, 100 mM NaCl, pH 5.0). a Structures of depsipeptides used to systematically characterize cationic side chain effects on RNA duplex stability. Lactic acid residues are highlighted. Ac acetyl, Aba acetamidobenzoic acid, which was appended to the N-terminus to increase UV absorbance. b Structures of the Fmoc-didepsipeptide building blocks used for solid-phase synthesis of depsipeptide sequences. Fmoc fluorenylmethoxycarbonyl. c Comparative effects of oligo-didepsipeptide sequences 2–6 and analogous oligo-dipeptide and on the T_m of the RNA duplex (number of independent measurements: n = 3 for Arg and Lys sequences; n = 2 for His and Dab sequences; n = 4 for Orn sequences). The shaded areas correspond to the mean ± SD T_m measured for the RNA duplex alone (n = 16 independent measurements). Data are shown as a scatter plot with mean ± SD. d Comparative effects on RNA duplex T_m of depsipeptide sequences containing a single backbone ester (number of independent measurements: n = 3 for Arg; n = 2 for His, Orn, and Dab; n = 4 for Lys), relative to RNA duplex alone (n = 16 independent measurements). Data are shown as a scatter plot with mean ± SD. e To assess the impact of different cationic side chains on depsipeptide degradation rates, depsipeptides 7–11 (40 μM) were incubated at 37 °C in buffer (100 mM HEPES, 10 mM NaCl, pH 7.3). After 5 min, the reactions were quenched by the addition of 3% TFA, and samples were analyzed by HPLC. Consistent with the hypothesis that sequences containing Orn or Dab adjacent to a backbone ester bond would undergo facile intramolecular O,N acyl transfer, only 37–39% of the starting depsipeptide remained, predominantly due to the formation of lactam products. In contrast, intramolecular degradation products were not observed for the sequences containing Arg, His, or Lys adjacent to the ester bond, and >80% of the intact depsipeptide were observed in these cases.

Fig. 4. Cationic depsipeptides directly interact with RNA.

a Structures of cationic peptide 12 and depsipeptide 13 used for gel mobility shift assays, and depsipeptide 14 used for circular dichroism studies. b Gel mobility shift assay with increasing concentrations (66.6–530 µM) of the FAM-labeled sequences 12 or 13 (green fluorescence) and a 5′-Cy5-U₂₀ RNA (26.6 µM, red fluorescence) in MES-TEA buffer (pH 6). A physical association between the RNA and cationic oligomers is evident as a less mobile band in the gel, which appears as orange due to co-localization of the green and red dyes, and as loss of intensity of the free RNA band. The gel was cropped at the edges for clarity. The image shown is representative of two independent experiments giving similar results. c CD spectra of RNA duplex 1 (5 μM each strand) with increasing concentrations of depsipeptide 14 (0–10 μM), indicating a concentration-dependent association. Spectra were recorded in 100 mM MES-TEA buffer (pH 6). d Plot of the change in CD signal at 266 nm as a function of increasing concentrations of depsipeptide 14. Red and blue colors of the filled circles correspond to curves of the same color in panel (c). Data were fit to a simple one-site binding model (black line) yielding an apparent K_d of ~3 μM.

Fig. 5. RNA increases the hydrolytic lifetime of a cationic depsipeptide.

a A schematic of depsipeptide-RNA interactions leading to increased depsipeptide lifetimes. The free depsipeptide is hydrolyzed with a pseudo-first rate constant, k_hyd-free. Binding of the depsipeptide to RNA is governed at equilibrium by k_assoc/k_dissoc. The pseudo-first rate constant for of depsipeptide hydrolysis within the depsipeptide–RNA complex is k_hyd-complex. Under conditions where complex formation is favorable and k_hyd-complex < k_hyd-free, the presence of RNA will increase the depsipeptide lifetime. b HPLC traces (270 nm) showing hydrolysis of depsipeptide 9 (25 μM) at various time points in the presence or absence of RNA duplex 1 at 37 °C in pH 7.3 buffer. The C-terminal fragment of the hydrolyzed depsipeptide is not observed because it lacks the Aba chromophore. (c) Time courses for hydrolysis of depsipeptide 9 (25 μM) with varying concentrations of the RNA duplex 1. The curves shown are from simultaneous fits of data to the model given in panel a using SimFit. During fitting, we fixed k_assoc = 1 × 10⁵ M⁻¹ s⁻¹. Therefore, three rate constants were fit, with values obtained by the fitting of: k_hyd-free = 1.1 × 10⁻⁴ s⁻¹, k_hyd-complex = 3.2 × 10⁻⁶ s⁻¹, and k_dissoc = 8.3 × 10⁻³ s⁻¹. d Kinetic profile of the hydrolysis reaction of depsipeptide 9 (25 μM) in the presence of 7 μM RNA duplex 1. Observed data are shown as filled circles, while the curves represent concentrations of molecular species predicted by SimFit modeling. e Comparison of depsipeptide hydrolysis in the absence of RNA, or in the presence of single-stranded RNA (100 μM 5′-rCrGrArUrUrUrArGrCrG-3′) or RNA duplex 1 (50 μM each strand). f To illustrate the mutually increased depsipeptide lifetime and RNA duplex T_m, three samples were prepared in parallel: one containing only RNA duplex 1 (25 μM each complementary strand), one containing only depsipeptide 9 (25 μM), and one containing both the RNA and 9 (at a 1:1 molar ratio). The extent of depsipeptide hydrolysis and RNA hybridization were then measured. Data are shown as a scatter plot of two independently repeated experiments.