An unnatural enzyme with endonuclease activity towards small non-coding RNAs

Abstract

Endonucleases are enzymes that cleave internal phosphodiester bonds within double-stranded DNA or RNA and are essential for biological functions. Herein, we use genetic code expansion to create an unnatural endonuclease that cleaves non-coding RNAs including short interfering RNA (siRNA) and microRNAs (miRNAs), a function that does not exist in nature. We introduce a metal-chelating unnatural amino acid, (2,2'-bipyridin-5-yl)alanine (BpyAla) to impart endonuclease activity to the viral suppressor of RNA silencing protein p19. Upon binding of copper, the mutant p19-T111BpyAla displays catalytic site-specific cleavage of siRNA and human miRNAs. Catalysis is confirmed using fluorescence polarization and fluorescence turn-on. Global miRNA profiling reveals that the engineered enzyme cleaves miRNAs in a human cell line. The therapeutic potential is demonstrated by targeting miR-122, a critical host factor for the hepatitis C virus (HCV). Unnatural endonuclease function is shown to deplete miR-122 levels with similar effects to an antagomir that reduces HCV levels therapeutically.

Fig. 1. Incorporation of BpyAla in 2 sites in the p19 dimer.

a Incorporation of BpyAla site-specifically into the VSRS protein p19 dimer introduces a metal binding site. Here we used copper to bind to BpyAla. If the metal can cleave the phosphodiester linkage within the p19 substrate, then it should induce site-specific strand cleavage of the bound siRNA or miRNA. Cleavage can occur on either strand. Created with BioRender.com. b Crystal structure of the p19 dimer with K67 (blue) and T111 (red) highlighted PDB: 1RPU. c Overall Schematic representation of the potential disruption of the RNA silencing pathway upon treatment with the engineered p19 endonuclease. Created with BioRender.com.

Fig. 2. siRNA cleavage by p19-T111BpyAla mutant.

a Binding plots for p19-T111BpyAla and for p19-K67BpyAla using electrophoretic mobility shift assay. Binding plots were constructed by varying the concentrations of the protein (0–1 µM), while maintaining the concentration of Cy3-labeled GL2 at 2 nM. b An overall schematic representation of the experimental strategy, in which Cy3-GL2 siRNA cleavage was assessed using a denaturing urea-PAGE gel. Created with BioRender.com. c 20% urea-PAGE gel was used to visualize the degradation product of Cy3-Gl2 siRNA by the engineered p19-T111BpyAla mutant in the presence of reducing agent and CuSO₄. Gel was independently reproduced 2 times, where different purified protein stocks were used.

Fig. 3. Detection of siRNA cleavage by p19-T111BpyAla using fluorescence polarization assay.

a The overall hydrolytic RNA cleavage mechanism due to the incorporation of BpyAla into p19. b An overall schematic representation of the experimental strategy, in which Cy3-Gl2 siRNA cleavage was assessed using fluorescence polarization. Created with BioRender.com. c Fluorescence polarization assay was used to visualize the cleaving potential of p19-T111BpyAla. The assay was conducted using 5 µM p19-T111BpyAla or p19-WT and 200 nM RNA, n = 3, in which n values represent 3 independent replicates. Red dots represent p19-T111BpyAla, and black dots represent p19-WT.

Fig. 4. Determination of Michaelis–Menten kinetic parameters using a fluorescence turn-on assay.

a A schematic representation of the enzymatic cleavage assay using a fluorescence turn-on approach, where if cleavage occurs an increase of Cy3 fluorescence is observed. Created with BioRender.com. b Steady-state cleavage velocities (MTO rates) are plotted against the concentration of substrate. 100 nM p19-T111BpyAla or p19-WT was incubated with varying concentrations of Cy3-GL2-BHQ-2 substrate to calculate the kinetic parameters at 37 °C. siRNA cleavage was monitored using fluorescence, in which initial velocities were derived from the initial slopes. Error bars represent mean values ± SD for 3 independent replicates, representing 3 independent purified protein batches. Black circles represent p19-T111BpyAla and Black triangles represent p19-WT.

Fig. 5. Cleavage of Huh7 miRNA samples using p19-T111BpyAla.

a Schematic representation of the workflow for small RNA isolation for Nanostring profiling and RT-qPCR analysis. RNA used was isolated from Huh7 hepatocellular carcinoma cell line and treated in the presence of CuSO₄. Created with BioRender.com. b A heat map depicting some of the significant downregulated fold changes in miRNAs upon treatment with p19-T111BpyAla relative to p19-WT treated miRNAs, n = 2, n represents 2 biological trials, where RNA is isolated from 2 independent cell passages. miRNAs with counts below 20 were eliminated from the heat map. Changes presented are more than 1.5-fold change. c qRT-PCR analysis of the relative miR-122 expression in p19-T111BpyAla treated samples in comparison to that treated with p19-WT in the presence of copper ions. Error bars represent the mean ± SD, n = 3, n values represent the treatment of RNA isolated from 3 different passages. d qRT-PCR analysis of the relative miR-122 expression in p19-T111BpyAla treated samples in the presence and absence of copper ions. Error bars represent the mean ± SD, n = 3, n values represent the treatment of RNA isolated from 3 different passages. e Derepression of miR-122’s downstream direct target STAT3 upon treatment with p19-T111BpyAla, Error bars represent the mean ± SD, n = 3. n values represent the transfection of cells from 3 different passages. Unpaired two-tailed t-test was used to evaluate statistical significance.

Fig. 6. p19-T111BpyAla effect on HCV replication.

a Schematic representation of the subgenomic HCV replicon (pFK-I389neo/luc/NS3-3′/5.1) used in this study and HCV JFH1_T strain. Created using BioRender.com. b Relative luciferase signal upon treatment of E9 cells with either miR-122 antagomir, c p19-T111BpyAla or p19-WT. The levels of miR-122 are directly correlated with HCV replication, which is linked to the relative luciferase signal in E9 cells. The effects of both miR-122 antagomir, and p19-T111BpyAla were investigated, with both treatments leading to a similar reduction in HCV subgenomic replication. For b and c, n = 3, and error bars represent mean values ± SD, p-values are presented. Unpaired two-tailed t-test was used to evaluate statistical significance. d Relative expression of HCV JFH1 IRES in Huh7.5 cells upon treatment with 1 µM p19-T111BpyAla, n = 3, error bars represent mean ± SD. P-values are presented in the figure. n values represent transfections of different cell passages. Control represents no treatment. Unpaired two-tailed t-test was used to evaluate statistical significance.