CN112877314B - Inducible base editing system and application thereof - Google Patents

Abstract

The invention discloses an inducible base editing system and application thereof. The base editing system comprises an inducible human-derived cytosine deaminase (A3A) and single-nicking activity SpCas 9; deaminase can split based on potential amino acid sites; rapamycin recombines into split deaminase by the interaction of FRB/FKBP with the N and C ends of the amino acid splitting site and induces base editing to be completed. The base editing activity of the induction type base editing system is controlled by constructing the induction type base editing system, so that off-target editing is reduced; at the same time, the split design in this application does not reduce target editing and therefore can be used as a compensation strategy for protein mutations and can be used in combination with these mutant deaminases to further reduce off-target editing.

Description

An inducible base editing system and its application

technical field

The invention belongs to the technical field of genetic engineering, and in particular relates to an inducible base editing system and its application.

Background technique

Base editing is a genome editing strategy that directly converts specific individual nucleotides in genomic DNA (Rees and Liu 2018). Currently, there are two types of base editors, cytosine base editors (CBE) and adenine base editors (ABE), both of which consist of a deaminase enzyme with a catalytically defective Cas9 protein that catalyzes Deamination of cytosine or adenine. Under the guidance of sgRNA, the base editor can convert C-G base pairs to T-A base pairs, or convert A-T base pairs to G-C base pairs within a few base pairs of a specific window within the target site. Because it is efficient and precise, and does not produce by-products such as insertions or deletions introduced by DNA double-strand breaks (DSBs), this technique was soon widely used in various model organisms, such as mice, zebrafish, simian Arabidopsis and Saccharomyces cerevisiae.

Although base editing technology has achieved many exciting results in a wide range of basic research, there is still a major obstacle in the clinical application of base editing, that is, off-target effects at the DNA or RNA level. Current CBEs and ABEs often cause transcriptome-wide off-targets in RNAs, raising safety concerns. More seriously, since CBE can also cause off-targets at the DNA level, it may further lead to pathological conditions including cancer. In fact, more than three decades ago, researchers discovered that overexpressing APOBEC1, one of the most popular cytosine deaminases used in CBE, caused liver cancer in transgenic mice. Subsequently, AID was also found to be overexpressed in mice to induce tumorigenesis. Based on these findings, an increasing number of studies have confirmed the link between cytosine deaminase and tumorigenesis in various tissues. Members of CDases, such as APOBEC3B, have been found in a variety of solid tumors.

Overexpression of cytosine deaminase has been found to induce base substitutions in genomic DNA, thereby compromising genomic stability. More importantly, many tumor cells have been found to harbor SNVs characterized by cytosine marks. Therefore, uncontrolled base editors, especially CBEs, may generate tumors. Considering that AAV is currently one of the most effective in vivo delivery vectors, AAV injection will cause long-term expression of transgenes in vivo, and the use of this vector to deliver base editors will undoubtedly amplify off-target consequences. Therefore, we urgently need to explore new base editing tools with controllable activity.

Contents of the invention

Aiming at the above-mentioned deficiencies in the prior art, the present invention provides an inducible base editing system and its application, by constructing a split base editing system, and controlling its base editing activity through the induction of rapamycin , the editing time can be expected to be shortened through this scheme. At the same time, the split design in this application does not reduce on-target editing, therefore, it can be used as a compensatory strategy for protein mutations and can be used in combination with these mutated deaminases to further reduce off-target editing.

In recent years, many studies have demonstrated the off-target effects of base editors on DNA and RNA. The invention controls the editing activity of the base editor by controlling the activity of the deaminase, thereby reducing the off-target effect. The base editing system of the present invention includes inducible human cytosine deaminase (A3A) and SpCas9 with single nick activity. The specific scheme is: after splitting A3A into inactive N-terminal and C-terminal, they are fused with FRB and FKBP proteins respectively, and FRB and FKBP form a stable ternary complex under the induction of rapamycin, so that the deaminase can be reactivated. Assemble into a functional A3A. Since the activity of the deaminase can be regulated by the inducer, the activity of the base editor is controllable.

In order to achieve the above object, the technical solution adopted by the present invention to solve the technical problems is:

An inducible base editing system, the base editing system includes a base editor, and rapamycin combined with it; the base editor includes a deaminase;

Deaminase can be split based on any amino acid site; rapamycin binds to deaminase through the split amino acid site and induces base editing.

Further, the base editor is a base editor composed of deaminase and BE3/SpCas9.

Further, FRB and FKBP of rapamycin are respectively coupled to the cleavage of the nonstructural loop of the deaminase amino acid cleavage site.

Further, FRB and FKBP were fused to the N-terminal and C-terminal of the deaminase amino acid splitting site, respectively, and dimerized to form a heterodimer to complete the assembly of the base editor.

Further, the splitting site of the deaminase is the 44th, 85th, 118th or 147th amino acid.

Further, the split site of the deaminase is the 85th amino acid.

Further, the deaminase is cytosine deaminase.

Further, the cytosine deaminase is APOBEC3A cytosine deaminase or APOBEC1 cytosine deaminase.

Further, the concentration of rapamycin is 0.01-200nM.

Further, the concentration of rapamycin is 200nM.

Application of the above-mentioned inducible base editing system in gene editing.

A kit for gene editing, comprising the above-mentioned inducible base editing system.

The base editing system constructed in this application can keep Cas9 intact, so it is compatible with other non-traditional base editors, such as CP-Cas9-derived base editors (Huang et al., 2019), internal mosaic BE - PIGS base editing (Wang et al., 2019) and sgBEs with sgRNA backbone modification (Wang et al., 2020).

The beneficial effects of the present invention are:

The present application constructs a split base editing system, and can control its base editing activity through the induction of rapamycin, and the editing time can be expected to be shortened through this scheme. At the same time, the split design in this application does not reduce on-target editing, therefore, it can be used as a compensatory strategy for protein mutations and can be used in combination with these mutated deaminases to further reduce off-target editing.

Description of drawings

Figure 1 is a flow chart of the design of the inducible split-A3A-BE3;

Figure 2 shows the results of inducible editing activity detection of 4 split-A3A-BE3;

Figure 3 is the detection results of the effect concentration and duration of rapamycin on the base editing of sA3A-BE3-85;

Figure 4 is the feature detection of the editing mode of sA3A-BE3-85;

Figure 5 shows the DNA off-target editing detection results of sA3A-BE3-85;

Figure 6 shows the detection results of reducing background editing by controlling the subcellular localization of sA3A-BE3-85 through the ERT2 system;

Figure 7 shows the construction process and feature detection of split-APOBEC1-BE3.

Detailed ways

The specific embodiments of the present invention are described below so that those skilled in the art can understand the present invention, but it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, as long as various changes Within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are included in the protection list.

Embodiment 1 constructs split-A3A plasmid

The split-A3A plasmid was constructed on the basis of A3A-BE3 to form a series of N-terminal split-A3A and its corresponding C-terminal split-BE3-A3As (Fig. 1b). As shown in Figure 1c, four groups of split-A3A-BE3 (sA3A-BE3) were constructed and named after their amino acid splitting sites, namely sA3A-BE3-44, sA3A-BE3-85, sA3A-BE3-118 and sA3A-BE3 were -147, respectively.

Figure 1 is the design of the inducible split-A3A-BE3; among them, panel a shows the schematic diagram of the reassembly of the inducible split-A3A-BE3. Dimerization to form a heterodimer completes the assembly of a functional A3A-BE3 base editor. Figure b is a schematic diagram of the structure of A3A (PDB: 5keg). The four unstructured loops opposite the ssDNA-A3A binding interface are candidate cleavage sites, represented as spheres in the figure. Figure c is a schematic diagram of the construction of the split-A3A-BE3 cytosine base editor.

Example 2 split-A3A base editing

1. Detection of inducible base editing activity of sA3A-BE3

Each pair of split-A3A-BE3 editors was separately co-transfected with sgRNAs for each of the 3 endogenous targets into HEK293T cells. As shown in Figure 2a, all sA3A-BE3s exhibited inducible editing activity, albeit with different potentials. Among these sA3A-BE3, sA3A-BE3-44 edited most efficiently at all 3 target sites when induced with 200 nM rapamycin, a concentration frequently used in the split system of FRB/FRBP. However, its activity remained essentially unchanged under uninduced conditions. On average, uninduced sA3A-BE3-44 showed comparable activity to full-length A3A-BE3 (Fig. 2b). Transfection of the C-terminal portion of sA3A-BE3-44 did not result in any detectable editing, suggesting that the background editing activity of sA3A-BE3-44 is not due to residual deamination activity of C-terminal A3A, but may be due to autologous deaminase activity. caused by assembly. In contrast, sA3A-BE3-147 exhibited minimal editing activity under both uninduced and induced conditions, which were about 7%-19% of the activity of full-length A3A-BE3, respectively (Fig. 2b). sA3A-BE3-85 and sA3A-BE3-118 were induced more efficiently than the above two. On average, sA3A-BE3-85 was similar in activity to full-length A3A-BE3 under inducing conditions, whereas sA3A-BE3-118 was approximately 1.7-fold less active than full-length A3A-BE3. Among the four split-A3A-BE3 base editors tested, sA3A-BE3-85 showed the best response to rapamycin (Fig. 2b).

Figure 2 shows the inducible editing activity of 4 split-A3A-BE3; among them, panel a is the assay of the editing efficiency of the inducible base editor, among the 3 endogenous targets, the full-length A3A-BE3 or the inducible Type sA3A-BE3s plasmids were co-transfected with sgRNA into HEK293T cells. Base editing efficiency was analyzed by Sanger sequencing and EditR (Kluesner et al. 2018). Each experiment was repeated at least three times. Data are expressed as mean ± SEM. Panel b shows the average editing efficiency of all editors in panel a for each target. The editing efficiency of full-length A3A-BE3 was normalized as 1.

2. Characterization of sA3A-BE3-85

The effect of rapamycin induction time on editing efficiency was examined. The above HEK293T cells transfected with sA3A-BE3-85 editor were treated with 200 nM rapamycin for 3, 6, 12, 24 and 48 hours, respectively. As shown in Fig. 3a,c, we observed time-dependent editing. Interestingly, of the three target sites tested, the editing efficiency of 24-hour induction was similar to that of 48-hour induction at two positions, while for the remaining one site, the editing efficiency of 24-hour induction was higher than that of 48-hour induction. 40.36% lower, suggesting that the effect of induction duration is target-specific. Next, we tested the effect of the concentration of rapamycin on editing efficiency by treating transfected cells with rapamycin concentrations ranging from 0.01 nM to 200 nM (Fig. 3b). As expected, a dose-dependent effect was observed, and the dose-response curve showed a plateau starting around 50 nM. (Fig. 3d).

Figure 3 is the effect of the concentration and duration of rapamycin on the base editing of sA3A-BE3-85; among them, Figure a is the editing efficiency of sA3A-BE3-85 at different induction times under 200nM rapamycin treatment . The transfected HEK293T cells were treated with 200 nM rapamycin, and the cells were harvested at corresponding time points. Panel b shows the editing efficiency of sA3A-BE3-85 induced by different concentrations of rapamycin for 48 hours. Transfected HEK293T cells were treated with various concentrations of rapamycin for 48 hours, followed by sequence analysis. The effects of induction duration (panel c) and intensity (panel d) on base editing in sA3A-BE3-85 were summarized by analyzing the average editing efficiency of all targetable cytosines at each site.

Further characterization of embodiment 3 split-A3A

To examine whether the cleavage of A3A at amino acid 85 changes the base editing mode of A3A-BE3, including editing window, sequence preference and editing purity of C→T.

To determine the editing window, 3 targets with multiple cytosines within the editing window were selected (Fig. 4a). As shown in Figure 4a, sA3A-BE3-85 did not significantly shift the position of the editing window or affect the width of the editing window compared to the full-length A3A-BE3. We then tested whether sA3A-BE3-85 has a different sequence preference than full-length A3A-BE3, as previous studies have shown that the unstructured loop between deaminase helices plays an important role in substrate recognition (Salter et al. 2016). The nine targets were divided into four groups according to NC motifs (GC, CC, AC, and TC), and the sequence preference of sA3A-BE3-85 and full-length A3A-BE3 was compared (Fig. 4b). On average, sA3A-BE3-85 edited at a similar efficiency to full-length A3A-BE3 in all four different motifs. sA3A-BE3-85 accounted for the relative editing efficiency of full-length A3A-BE3, which was 84.5% for GC, 99.7% for CC, 93.5% for TC, and 95.0% for AC, which indicated that sA3A-BE3-85 There were no significantly altered motif preferences. To test the purity of C to T editing, we selected 2 previously characterized sites that, in addition to C to T, also tend to switch from C to G (C6 in HEK2 and C6 in RNF2). We found that the edited purity of sA3A-BE3-85 was 53.05% higher than the original A3A-BE3 and 9.18% higher at the RNF2 site, which indicated that sA3A-BE3-85 slightly improved the purity of C→T products with higher C→T Efficiency (Fig. 4c).

Figure 4 shows the characteristics of the editing mode of sA3A-BE3-85; among them, Figure a is the comparison of the editing window between sA3A-BE3-85 and full-length A3A-BE3; Figure b is the comparison between sA3A-BE3-85 and full-length A3A-BE3 Sequence preference comparison, calculating the base editing efficiency of various NC motifs in each target; Figure c shows the comparison of the editing purity of full-length A3A-BE3 and sA3A-BE3-85.

Example 4 DNA off-target editing of split-A3A-BEs

The fact that CBE induces both sequence-dependent and sequence-independent off-target editing greatly limits its application. To determine the sequence-dependent off-target editing of sA3A-BE3-85, two well-characterized targets (HEK site4 and EXMI) were selected. As shown in Figure 5a, sA3A-BE3-85 generally showed lower sequence-dependent off-target editing at all three off-target sites. The editing efficiency of sA3A-BE3-85 at off-target sites gradually increased with the induction time (Fig. 5a). The ratio of editing efficiency to off-target efficiency was calculated and found that sA3A BE3-85 (4:1) was higher than A3A-BE3 (2.4:1) (Fig. 5b).

The sequence-independent off-target editing of sA3A-BE3-85 was then tested by an orthogonal R-loop assay, and artificial R-loops were induced by transfection of the single nickase saCas9 and sgRNA at specific genomic loci. Co-transfection of base editors with R-loop constructs revealed that sA3A-BE3-85 displayed a significant reduction in off-target editing compared to full-length A3A-BE3 (Fig. 5c). The above results indicated that sA3A-BE3-85 showed a lower propensity for both sequence-dependent and sequence-independent DNA off-target editing compared with full-length A3A-BE3.

Figure 5 shows the DNA off-target editing of sA3A-BE3-85, in which, Figure a shows the sequence-dependent off-target editing efficiency of sA3A-BE3-85, and two gRNAs targeting HEK4 and EMX1 sites were used to analyze sequence-dependent off-target Edit, off-target sites were predicted by sequence similarity and have been described in (Komor et al. 2016). The average editing efficiency of all targetable cytosines was analyzed at each site; panel b is the relative editing efficiency calculated from the ratio of on-target editing to off-target editing; panel c shows the principle of detection by orthogonal R loop and sA3A-BE3-85 Sequence-independent detection of off-target editing. Two artificial R-loops were generated by transfecting nsaCas9 and the corresponding sgRNA, and their target editing efficiency was measured simultaneously (EXM1)

Example 5 Reduces background editing of split-A3A by controlling subcellular localization

Since split-A3A-85 exhibits background editing when not induced by rapamycin, we sought to test whether background editing could be further reduced by modulating the subcellular localization of split-A3A-85 components so that Spatially separated upon induction.

The nuclear transport system ERT2 was used to control nucleic acid transport of the N-terminal portion of split-A3A-85, so that the N-terminal split-A3A-85 was localized in the cytoplasm and sterically separated from its C-terminus, thereby blocking automatic assembly (Fig. 6a). As shown in Figure 6b, fusion of the ERT2 domain to the N-terminus but not the C-terminus of N-split-A3A-85 achieved inducible editing under induction and significantly reduced background editing without induction.

Figure 6 Controlling the subcellular localization of sA3A-BE3-85 with the ERT2 system reduces background editing; among them, panel a shows the mechanism for controlling the nuclear translocation of sA3A-BE3-85 with the ERT2 system, the N-terminal part of sA3A-BE3-85 Fused with ERT2 and induced by 4-OHT to transfer into the nucleus, where it is close to the C-terminal part of sA3A-BE3-85, and assembles a functional base editor in the presence of rapamycin; Figure b is Construction of the sA3A-BE3-85 construct induced by the dual control system of 4-OHT and rapamycin; panel c is the background editing assay of split-A3A-BE3-85 fused to ERT2.

Example 6 Design and Characterization of Base Editors Derived from Rat APOBEC1

Having established the split-A3A-BE3 system, we next sought to extend this finding to another widely used cytosine deaminase, rAPOBEC1, which is also frequently used in base editing. Through sequence alignment, rAPOBEC1 has 48.1% similarity with A3A, and the key amino acids responsible for catalytic deamination are basically conserved between these two cytosine deaminases. Since the crystal structure of rAPOBEC1 has not been resolved, we used an online program (MPI BioinformaticsToolkit: https://toolkit.tuebingen.mpg.de.) to predict its structure from the protein sequence. The predicted structure of rAPOBEC1 is similar to that of A3A, especially the loop 4 and nearby structures in A3A are highly homologous to the corresponding regions in rAPOBEC1. As shown in Fig. 7a,b, we mapped the 85th amino acid of A3A to the 77th amino acid of rAPOBEC1 by structure comparison and sequence alignment. The resulting split-rAPOBEC1-BE3 achieved 85.83% of the activity of full-length BE3 under inducing conditions, while its activity was significantly reduced without disturbance, which is consistent with the previous results of sA3A-BE3-85 (Fig. 7c). Therefore, this result suggests that our resolution strategy can also be applied to other deaminases.

Figure 7 shows the design and characteristics of split-APOBEC1-BE3; among them, Figure a is the amino acid sequence alignment of APOBEC3A and APOBEC1; Figure b is the structure comparison of A3A and APOBEC1, and the MPI bioinformatics toolkit predicted the structure of APOBEC1, red The region shows the split site; panel c is the detection of the induction activity of split-APOBEC1-BE3.

The present application constructs a split base editing system, and can control its base editing activity through the induction of rapamycin, and the editing time can be expected to be shortened through this scheme. At the same time, the split design in this application does not reduce on-target editing, therefore, it can be used as a compensatory strategy for protein mutations and can be used in combination with these mutated deaminases to further reduce off-target editing.

In addition, the base editing system constructed in this application can keep Cas9 intact, so it is compatible with other non-traditional base editors, such as CP-Cas9-derived base editors (Huang et al., 2019) , internal mosaic BE-PIGS base editors (Wang et al., 2019) and sgBEs with sgRNA backbone modification (Wang et al., 2020).

Claims (6)

1. An inducible base editing system, characterized in that, the base editing system includes a base editor and rapamycin combined therewith; the base editor includes a split deaminase; The deaminase is human cytosine deaminase APOBEC3A with a sequence length of 199 amino acids; The human cytosine deaminase APOBEC3A is split at the 85th amino acid position, and the N-terminal and C-terminal of the split site are respectively fused with FRB and FKBP proteins, and FRB and FKBP can be induced by rapamycin Dimerization forms a heterodimer, thereby assembling an active human cytosine deaminase APOBEC3A. 2. The inducible base editing system according to claim 1, wherein the FRB and FKBP are respectively coupled to the cleavage of the nonstructural loop of the deaminase amino acid splitting site. 3. The inducible base editing system according to claim 1, wherein the deaminase is replaced by rat cytosine deaminase APOBEC1 with a sequence length of 229 amino acids; the rat cytosine deaminase The ammoniaase APOBEC1 was resolved at the 77th amino acid position. 4. The inducible base editing system according to claim 1, wherein the concentration of rapamycin is 0.01-200nM. 5. Application of the inducible base editing system according to any one of claims 1 to 4 in gene editing. 6. A kit for gene editing, comprising the inducible base editing system according to any one of claims 1 to 4.

Images