Photosensitive RNA cytidine base editor and use thereof

Abstract

The present invention relates to a photosensitive RNA cytidine base editor and use thereof. The photosensitive RNA cytidine base editor can cause deamination of cytidine (C) in a target double-stranded RNA to uridine (U), thereby achieving editing of C to U. The photosensitive RNA cytidine base editor comprises: (1) a dCas-eADAR2dd N-pMagHigh fusion protein, and (2) an nMagHigh-eADAR2dd C fusion protein.

Description

A photosensitive RNA cytidine base editor and its application Technical Field

[0001] This invention belongs to the field of biotechnology, specifically relating to a photosensitive RNA cytidine base editor and its applications. Background Technology

[0002] Site-specific RNA base editing is an emerging gene modification technology that enables precise conversion of specific bases directly within RNA, showing great potential for disease treatment. Unlike DNA editing, which has two editing states—"on" and "off"—RNA base editing does not alter the genome sequence, can flexibly regulate the expression levels of target proteins, and has good reversibility. These characteristics make RNA base editing a safer and more flexible therapeutic approach, mitigating some of the ethical issues associated with genome modification. RNA base editors capable of effective editing in mammals fall into two categories: the former can achieve A (adenosine) > I (inosine) base conversion on RNA, while the latter can achieve C (cytidine) > U (uridine) base conversion. Recently, researchers have developed a series of small, highly efficient, and specific A > I RNA base editing tools. However, the development of C > U RNA base editing has been relatively slow, and the performance of existing base editors urgently needs improvement.

[0003] Based on the source of the cytidine deaminase in the editor, C>U RNA base editing tools can be divided into two main categories. The first category uses cytidine deaminases from the APOBEC family as the catalytic core, such as the CURE and CU-REWIRE systems. CURE consists of dCasRX and APOBEC3A. This system has high specificity, but it has a strict preference for 5'-UCR (R=A,G). 1 CU-REWIRE is a gRNA-independent editing system composed of the RNA-specific binding protein PUF and APOBEC3A. 2 This system can achieve efficient editing in mice, but its specificity is poor. Although both CURE and CU-REWIRE can achieve pure and single C>U editing, their core catalytic enzyme APOBEC3A has potential off-target effects at both the DNA and RNA levels. In 2019, Zhang Feng's team obtained an evolved ADAR mutant with C>U editing activity, abbreviated as eADAR, by screening for evolved adenine deaminases, and fused it with dCas13b to construct the RESCUE system. 3 Subsequently, the researchers fused eADAR with various small Cas proteins to construct a series of novel C>URNA base editors, such as xCBE, mxCBE, RESCUE.t1, and RESCUE.t3.4,5 While both of the aforementioned RNA base editors can achieve effective editing, persistent expression of cytidine deaminases can lead to significant off-target editing. Furthermore, these RNA base editors lack spatiotemporal specificity, making them difficult to use for regulating biological processes based on temporal and spatial gene expression patterns, especially in animals. Recently, the Won Do Heo team in South Korea constructed a blue light-activated C-to-U RNA base editor (padCas13 editor) by fusing a Magnets system-driven recombinant segmented dCas13b protein with the full-length eADAR2dd (evolved deaminase domain of adenosine deaminase acting on RNA type 2) to achieve spatiotemporally specific C-to-U base editing in cells. 8 Similar to previous reports on the splitting of the Cas13 protein,6,7 the segmented dCas13b protein also exhibited significant self-assembly activity, leading to significant leakage activity of the padCas13 editor in the absence of external stimulation. Furthermore, the relatively large size of dCas13b (1127aa) affects the delivery efficiency of the padCas13 editor when using AAV as a vector. Therefore, there is an urgent need to develop a more precise, efficient, small-sized C>U RNA base editing tool with high spatiotemporal specificity.

[0004] Based on this, the present invention is proposed.

[0005] [References]

[0006] 1.Huang,X.et al.Programmable C-to-U RNA editing using the human APOBEC 3A deaminase.EMBO J.39,1–12(2020).

[0007] 2. Han, W. et al. Programmable RNA base editing with a single gRNA-free enzyme. Nucleic Acids Res. 50, 9580-9595 (2022).

[0008] 3. Abudayyeh, OO et al. A cytosine deaminase for programmable single-base RNA editing. Science (80-.). 365, 382-386 (2019).

[0009] 4. Xu, C. et al. Programmable RNA editing with compact CRISPR-Cas13 systems from uncultivated microbes. Nat. Methods 18, 499-506 (2021).

[0010] 5. Kannan, S. et al. Compact RNA editors with small Cas13 proteins. Nat. Biotechnol. 40, 194-197 (2022).

[0011] 6. Xu, Y. et al. A Split CRISPR / Cas13b System for Conditional RNA Regulation and Editing. J. Am. Chem. Soc. 145, 5561-5569 (2023).

[0012] 7. Ding, Y., Tous, C., Choi, J., Chen, J. & Wong, WW Orthogonal inducible control of Cas13 circuits enables programmable RNA regulation in mammalian cells. Nat. Commun. 15, 1572 (2024).

[0013] 8. Yu, J. et al. Programmable RNA base editing with photoactivatable CRISPR-Cas13. Nat. Commun. 15, 1-14 (2024).

[0014] Summary of the Invention

[0015] This invention first relates to a photoactivatable C-to-U RNA base editor (PA-rCBE), which can cause cytidine (C) in the target double-stranded RNA to deaminate into uridine (U), thereby achieving C>U editing;

[0016] The light-sensitive RNA cytidine base editor includes:

[0017] (1)dCas-eADAR2ddN -pMagHigh fusion protein, consisting of dCas protein and an N-terminal truncated fragment of the evolved adenine deaminase catalytic domain (eADAR2dd). N The pMagHigh monomer of the Magnets system, which is induced by blue light dimerization;

[0018] (2)nMagHigh-eADAR2dd C The fusion protein consists of an nMagHigh monomer of the Magnets system, dimerized by blue light, and a C-terminal truncated fragment of the catalytic domain of an evolved adenine deaminase (eADAR2dd). C )constitute;

[0019] The pMagHigh and nMagHigh can be induced by blue light to dimerize into the Magnets system (see PMID:25708714), thereby driving the C-terminus and N-terminus of the segmented eADAR2dd protein to recombine into an active adenine deaminase, which catalyzes the deamination of cytidine (C) of the target RNA into uridine (U).

[0020] The dCas protein is used to bind target RNA. The dCas protein includes, but is not limited to, mini dCas13X, dCasRx, dCas13X or EcCas6e protein, preferably the mini dCas13X protein with a smaller molecular weight.

[0021] Preferably, the sequence of the eADAR2dd is as shown in SEQ ID NO.1; the eADAR2dd C The sequence is shown in SEQ ID NO.2; the eADAR2dd N The sequence is shown in SEQ ID NO.3;

[0022] The sequence of pMagHigh is shown in SEQ ID NO.4; the sequence of nMagHigh is shown in SEQ ID NO.5;

[0023] The sequence of the mini dCas13X is shown in SEQ ID NO.6.

[0024] Furthermore, the various functional blocks of the fusion protein are linked together by a linker; the linker is:

[0025] (1) Connect eADAR2dd N The GS-type linker for the dCas protein, preferably, has the sequence shown in SEQ ID NO.7 (L30);

[0026] (2) Connect eADAR2dd N The linkerN between pMagHigh, preferably, has the sequence shown in SEQ ID NO.8 (Linker53);

[0027] (3) Connect eADAR2dd C The linker C between nMagHigh, preferably, has the sequence shown in SEQ ID NO.9;

[0028] Furthermore, the blue light induction condition is: 0.25 mW / cm². 2 Turn on in 1 minute and turn off in 3 minutes. Blue light wavelength is 470nm.

[0029] This invention also relates to the following applications of the aforementioned photosensitive RNA cytidine base editor:

[0030] (1) Targeted editing of eukaryotic or prokaryotic cells;

[0031] (2) To prepare drugs for treating diseases caused by gene mutations;

[0032] (3) Used for crop genetic breeding;

[0033] (4) Used to prepare animal models.

[0034] The beneficial effects of this invention are as follows:

[0035] (1) A photo-induced segmented eADAR protein dimerization strategy was adopted, instead of relying on segmented Cas13 protein recombination. We fused the inactive eADAR2dd fragments with a blue light-induced dimerization Magnets system for expression, constructing a blue light-activated PA-rCBE (photoactivatable C-to-U RNA base editor) system using a smaller mini dCas13X protein as an RNA binding platform. Only after blue light stimulation could the dimerization of pMag and nMag lead to the recovery of deaminase activity after the recombination of the segmented eADAR2dd fragments, thereby achieving C-to-U editing at the RNA level.

[0036] (2) To obtain an RNA base editor with high editing efficiency and no leakage, the length and amino acid composition of the linker sequence in the PA-rCBE system were systematically optimized. The results showed that the optimized PA-rCBE could respond sensitively to blue light, and the induction fold was as high as 132-fold when using luciferase as a reporter system. This system can not only achieve spatiotemporally specific RNA base editing in cells, but also sensitively regulate the intensity of editing activity by adjusting the time and intensity of light.

[0037] (3) Unlike the padCas13 editor, PA-rCBE directly regulates eADAR2dd activity, which can not only achieve spatiotemporally specific RNA base editing, but also achieve strict supervision of eADAR2dd activity. It has almost no off-target effects in the transcriptome and shows high fidelity.

[0038] In summary, the PA-rCBE system is a novel RNA base editing tool that integrates high-efficiency editing, low off-target effects, and precise control, laying a technological foundation for the precise manipulation of important physiological and pathological processes from an RNA perspective. Attached Figure Description

[0039] Figure 1. Construction of the PA-rCBE system. (a) Working principle of PA-rCBE. (b) Schematic diagram of evaluating RNA base editing efficiency using luciferase (Gluc C82R) as a reporter. (c) Screening for light-sensitive proteins for RNA base editing. (d) dCas-eADARdd N Four candidate vectors for -pMagHigh and nMagHigh-eADARdd C Schematic diagram of the plasmid. (e) Test dCas-eADARdd N Luciferase retrieval efficiency of the -pMagHigh candidate vector. Data are expressed as mean ± SD, n = 3.

[0040] Figure 2. Optimized PA-rCBE system. (a)eADAR2dd N A schematic diagram of the linker sequence between the mini dCas13X and the dCas13X. (b)eADAR2dd N Optimization of the linker sequence between the mini dCas13X and the mini dCas13X. (c)eADAR2dd N Flowchart of linker library construction between pMagHigh. (d) Library construction and screening using a degenerate primer strategy for eADAR2dd. N The linker between pMagHigh and the data. Data is expressed as mean ± sd, n = 3.

[0041] Figure 3. Editing of endogenous transcripts using PA-rCBE. (a) Editing efficiency of PA-rCBE, mxCBE, and RESCUE-S for various endogenous transcripts. (b) Average editing efficiency of each RNA base editor for eight endogenous transcripts. (c) Bystander off-target effects of each base editor on the CTNNB1 gene. (d) Bystander off-target effects of each base editor on the SMARCA4 gene. (e) Quantitative analysis of the number of off-target sites for the PA-rCBE and RESCUE-S systems under both CTNNB1-targeted and non-targeted (NT) conditions. EGFP was used as a negative control. All experiments were performed in HEK293T cells, and editing efficiency was quantified by NGS. Data are expressed as mean ± SD, n = 3.

[0042] Figure 4. Characterization of PA-rCBE. (a) Dose-dependent PA-rCBE editing activity. (b) Light-time-dependent PA-rCBE editing activity. (c) Decrease in editing effect over time after light cessation. (d) PA-rCBE reversibility test. (e) Comparison of PA-rCBE editing efficiency using a single target and tandem dual targets. Data are expressed as mean ± SD, n = 3. Detailed Implementation

[0043] The sequence structure of each structural unit of the PA-rCBE construct is as follows:

[0044] eADAR2dd, the sequence is shown in SEQ ID NO.1; SEQ ID NO.1:

[0045] eADAR2dd N The sequence is shown in SEQ ID NO.2; SEQ ID NO.2:

[0046] eADAR2dd C The sequence is shown in SEQ ID NO.3; SEQ ID NO.3:

[0047] pMagHigh, sequence as shown in SEQ ID NO.4; SEQ ID NO.4:

[0048] nMagHigh, sequence as shown in SEQ ID NO.5; SEQ ID NO.5:

[0049] mini dCas13X, sequence as shown in SEQ ID NO.6; SEQ ID NO.6:

[0050] The linker sequences are as follows:

[0051] L30, the sequence is shown in SEQ ID NO.7; SEQ ID NO.7: GGGGGSGGGGSGGGGSGGGGSGGGSGGGS

[0052] Linker N, sequence as shown in SEQ ID NO.8; SEQ ID NO.8; TVKRTSRA

[0053] Linker C, sequence as shown in SEQ ID NO.9; SEQ ID NO.9: GSGGGG

[0054] Material:

[0055] 1. CMV, U6, DR, dCasRx, dCas13X or EcCas6e are all conventional components in the art and are well known to those skilled in the art.

[0056] 2. Gluc C82R System: To facilitate rapid testing of PA-rCBE editing efficiency, we constructed a Gaussian luciferase reporter system carrying an inactivation mutation, abbreviated as Gluc C82R. Site-specific targeting of the 82nd amino acid of the Gluc C82R transcript converts the CGC encoding arginine to the UGC encoding cysteine, thereby functionally restoring the activity of Gaussian luciferase. The editing efficiency of PA-rCBE can be determined by detecting the intensity of luciferase. Specifically, the PA-rCBE candidate plasmid and the Gluc C82R reporter plasmid were co-transfected into HEK-293T cells. Twelve hours after transfection, one group of cells was irradiated with blue light (0.25 mW / cm²). 2 One group was placed in the light-exposed group (on for 1 minute, off for 3 minutes), while the other group was placed in the dark for protection from light. Forty-eight hours after transfection, the cell supernatant was collected to detect luciferase activity. Theoretically, the higher the fluorescence value, the stronger the PA-rCBE activity. The ratio of the fluorescence value of the light-exposed group to that of the dark-exposed group was used as the PA-rCBE induction fold; a higher value indicated a stronger light dependence of PA-rCBE.

[0057] Example 1: Construction of a blue light-activated C>U RNA base editing system (PA-rCBE)

[0058] To construct a photoactivatable C-to-U RNA base editor (PA-rCBE), we will use a segmented inactivated eADAR2dd fragment (eADAR2dd...). Nand eADAR2dd C These proteins are fused with different light-sensitive proteins for expression. Only under light stimulation can the light-sensitive proteins dimerize, leading to the recombination of the segmented eADAR2dd fragments, thus restoring deaminase activity and enabling C-to-U editing at the RNA level (Figures 1a and 1b).

[0059] We selected several groups of relatively small-sized dimerized photosensitive proteins to induce conditional recombination of segmented eADAR. These are the red light-induced MagRed system, the red light-induced nanoReD system, and the blue light-induced Magnets system. (Information on various photosensitive proteins can be found at PMID:35697806, PMID:33179507, PMID:25708714). Among them, Aff6-DrBphP and 2×Aff6-DrBphP are two combinations of the MagRed system. nMagHigh-pMag and nMagHigh-pMagHigh are strong dimerization variants of Magnets.

[0060] The aforementioned photosensitive proteins are respectively related to eADAR2dd N and eADAR2dd C Different versions of the PA-rCBE system were constructed by fusion expression and co-transfected with the reporter system Gluc C82R into HEK293T cells. The results showed that PA-rCBE had the highest editing efficiency when pMagHigh and nMagHigh were used as light-controlled switches (Figure 1c).

[0061] We combined four catalytically inactivated Cas proteins (dCasRx, dCas13X, EcCas6e, or mini dCas13X) with eADAR2dd. N -pMagHigh co-expression was used to construct different versions of dCas-eADAR2dd N -pMagHigh fusion protein and combined with nMagHigh-eADAR2dd C HEK293T cells were co-transfected with the report system (Fig. 1d).

[0062] The results showed that the PA-rCBE system constructed using mini dCas13X as the RNA recruitment platform had the highest editing efficiency (Figure 1e).

[0063] Example 2: Obtaining a high-efficiency and low-leakage PA-rCBE through molecular optimization

[0064] The linker sequence between gene elements has a significant impact on the activity of fusion proteins.

[0065] To improve the editing performance of PA-rCBE, we cloned GS linkers of different lengths into eADAR2dd. N Several PA-rCBE variants were obtained between the mini dCas13X and the mini dCas13X (Fig. 2a).

[0066] The results showed that PA-rCBE with L30 as the linker had the highest activity, with a photoinduction factor of up to 70 times (Figure 2b). The sequence of L30 is shown in SEQ ID NO.7.

[0067] Next, further optimization of eADAR2dd N To obtain a richer array of linker sequences between the linker and pMagHigh, we constructed PA-rCBE libraries containing different linkers using a degenerate primer strategy (Figure 2c). Luciferase assay results showed that PA-rCBE containing linker 53 exhibited the highest activity, with a photoinduction fold of 132-fold (Figure 2d). Linker 53 (linker N) has the sequence shown in SEQ ID NO. 8.

[0068] The different linker numbers and amino acids are shown in the table below.

[0069] Similarly, eADAR2dd C The best test result for the linker between nMagHigh and linker C is linker C (linker GS), with the sequence shown in SEQ ID NO.9.

[0070] In summary, through a series of molecular optimizations, we obtained a light-activated C-to-U RNA base editor with high editing efficiency and almost no background (leakage expression under dark conditions).

[0071] Example 3: Testing the editing efficiency and specificity of PA-rCBE in endogenous genes.

[0072] While reporting systems can rapidly characterize the features of PA-rCBE, their applicability is limited and they cannot accurately reflect the system's editing efficiency. To further test the editing efficiency of PA-rCBE, we selected eight endogenous transcripts with different functions and designed targets, systematically comparing the editing efficiencies of PA-rCBE, mxCBE, and RESCUE-S.

[0073] To evaluate the editing efficiency of PA-rCBE, we fused IRES-EGFP to the C-terminus of PA-rCBE as a flow cytometry sorting tag. 70,000 EGFP-positive cells were sorted from the transfected samples using a flow cytometer, and RNA was extracted.

[0074] The results show:

[0075] (1) PA-rCBE outperformed mxCBE in editing efficiency in 8 endogenous transcripts and was comparable to RESCUE-S. Its average editing rate was 37.2%, while that of mxCBE was 11.7% and that of RESCUE-S was 38.1% (Figure 3a, b).

[0076] (2) PA-rCBE showed virtually no bystander off-target effects of C-to-U, demonstrating its higher specificity (Fig. 3c, d).

[0077] To evaluate the specificity of PA-rCBE, we compared it with the previously reported high-fidelity RESCUE-S. The specific procedure was as follows: For samples used to detect the off-target effects of PA-rCBE in the transcriptome, approximately 600,000 EGFP-positive cells were collected and RNA extracted. The concentration and purity of each RNA sample were measured, and after quality control, a library was constructed using PolyA to enrich the RNA for transcriptome sequencing. Samples transfected only with EGFP served as the control group, and the editing events generated in this group were used as the background. After subtracting the background from the editing events in each experimental group, the total number of off-target events of PA-rCBE in the transcriptome was obtained.

[0078] Whole transcriptome RNA sequencing showed that PA-rCBE-induced off-target events were much lower than RESCUE-S (Figure 3e), demonstrating that PA-rCBE exhibits high-fidelity activity across the transcriptome.

[0079] Example 4: Characterization of PA-rCBE

[0080] (1) The characteristics of PA-rCBE were fully characterized using endogenous transcripts. Similar to existing RNA base editors, PA-rCBE exhibits good dose-dependent behavior, with editing efficiency gradually increasing with increasing transfection amount (Figure 4a).

[0081] (2) To further characterize the light dependence of PA-rCBE, we treated PA-rCBE-transfected cells with light for different durations. NGS sequencing results showed that PA-rCBE responded sensitively to blue light, and its editing efficiency increased with increasing light exposure time (Figure 4b). After the light exposure stopped, RNA base editing activity was shut down, and a significant decrease in editing effect could be observed after 24 hours (Figure 4c). Re-examination with light restored the editing activity of PA-rCBE, thus achieving reversible RNA editing (Figure 4d).

[0082] (3) Considering that CRISPR-Cas13 has the ability to self-process and cut RNA, we expressed multiple endogenous targets in tandem and co-transfected them with PA-rCBE into HEK293T cells. High-throughput sequencing (NGS) results showed that PA-rCBE still had good editing efficiency when using tandem dual targets, proving that PA-rCBE has the function of multi-pathway synergy (Figure 4e).

[0083] In summary, PA-rCBE can respond sensitively to blue light, rapidly and reversibly regulate gene expression, and can be used in combination with multiple pathways.

[0084] Finally, it should be noted that the above embodiments are only used to help those skilled in the art understand the essence of the present invention, and are not intended to limit the scope of protection of the present invention.

Claims

1. A photoactivatable RNA base editor (PA-rCBE), wherein the photoactivatable RNA base editor can induce the deamination of cytidine (C) in the target double-stranded RNA to uridine (U), thereby achieving C>U editing; The light-sensitive RNA cytidine base editor includes: (1)dCas-eADAR2dd N -pMagHigh fusion protein, consisting of dCas protein and an N-terminal truncated fragment of the evolved adenine deaminase catalytic domain (eADAR2dd). N The pMagHigh monomer of the Magnets system, which is induced by blue light dimerization; (2)nMagHigh-eADAR2dd C The fusion protein consists of an nMagHigh monomer of the Magnets system, dimerized by blue light, and a C-terminal truncated fragment of the catalytic domain of the evolved adenine deaminase (eADAR2dd). C )constitute; The pMagHigh and nMagHigh can be induced to dimerize into a Magnets system by blue light.

2. The photosensitive RNA cytidine base editor according to claim 1, characterized in that, The dCas proteins include, but are not limited to, mini dCas13X, dCasRx, dCas13X or EcCas6e proteins, preferably mini dCas13X proteins.

3. The photosensitive RNA cytidine base editor according to claim 1 or 2, characterized in that, The sequence of eADAR2dd is shown in SEQ ID NO.1; the eADAR2dd C The sequence is shown in SEQ ID NO.2; the eADAR2dd N The sequence is shown in SEQ ID NO.3; The sequence of pMagHigh is shown in SEQ ID NO.4; the sequence of nMagHigh is shown in SEQ ID NO.5; The sequence of the mini dCas13X is shown in SEQ ID NO.

6.

4. The photosensitive RNA cytidine base editor according to any one of claims 1-3, characterized in that, The functional blocks of the fusion protein are linked together by a linker; the linker is: (1) Connect eADAR2dd N The GS-type linker for the dCas protein, preferably, has the sequence shown in SEQ ID NO.7 (L30); (2) Connect eADAR2dd N The linkerN between pMagHigh, preferably, has the sequence shown in SEQ ID NO.8 (linker53); (3) Connect eADAR2dd C The linker C between nMagHigh, preferably, has the sequence shown in SEQ ID NO.9 (linker GS); 5. The light-activated RNA cytidine base editor of any one of claims 1-4, wherein The blue light induction condition is: 0.25 mW / cm². 2 Turn on in 1 minute and turn off in 3 minutes. Blue light wavelength is 470nm.

6. The following applications of the photosensitive RNA cytidine base editor according to any one of claims 1-5: (1) Targeted editing of eukaryotic or prokaryotic cells; (2) To prepare drugs for treating diseases caused by gene mutations; (3) Used for crop genetic breeding; (4) Used to prepare animal models.

Images