A nuclease and editing system with novel base editing function

Abstract

The present application provides an enzyme Ddd_SS from Simiaoa sunii, which can effectively remove the amino group of cytidine in GC environment. The present application further provides an optimized derivative DdCBE of Ddd_SS. Using the enzymes provided by the present application, gene editing can be carried out in GC environment, thereby introducing disease-related mitochondrial DNA (mtDNA) mutations, which cannot be achieved in the prior art.

Description

Background Technology

[0001] Base editing enables precise mutations in target DNA sequences, aiding in the analysis of gene and regulatory element function, disease modeling, and the development of therapeutic drugs. Base editing in nuclear DNA is achieved through base editing systems based on clustered regularly interspaced short palindromic repeats (CRISPR). Unfortunately, CRISPR base editing systems are currently unsuitable for mitochondrial DNA editing due to the lack of methods for delivering guide RNA to mitochondria. Recently, DddA-derived Cytosine Base Editors (DdCBEs), derived from Transcription Activator Like Effector (TALE), have been developed to catalyze C→T editing in mitochondrial DNA. These methods rely on DddAtox, a double-stranded DNA editing enzyme from Burkholderia cenocepacia. The original DddAtox requires a strictly defined TC sequence site. Through phage-assisted evolution, Mok et al. further obtained a DddAtox variant that can catalyze C→T editing at the TC / AC / CC sequence site. However, a suitable DdCBE for the GC sequence site still does not exist. Furthermore, it is unclear whether double-stranded DNA deaminases have been found in other species. In this study, the inventors identified and designed novel double-stranded DNA deaminase homologues to address these issues. Summary of the Invention Attached Figure Description

[0002] Figure 1 This demonstrates that the C-terminal sequence of the DddAtox protein plays a crucial role in its dsDNA deamination activity. Figure 1 (a): Boxes indicate SPKK motifs in the reported histone H2B.1 protein sequence of sea urchin (Psammechinus miliaris) and SPKK motifs in the DddA protein sequence of Burkholderia cenocepacia. Figure 1 (b): Schematic diagram of the protein structure of wild-type DddAtox (WT), a truncated form of DddAtox with a similar SPKK motif deleted (Delta), and an AT-Hook sequence fused to the truncated form (AT-Hook). Figure 1 (c): will Figure 1(b) The proteins shown were mixed with an equal amount of substrate dsDNA at different concentrations to determine the deamination activity of the proteins by identifying the relative proportion of dsDNA that underwent deamination to the total amount of substrate. The data shown in the figure are mean ± standard deviation, and the experiment was repeated 3 times independently. Figure 2 A comparison of various dsDNA cytosine deaminases homologous to DddAtox is shown. Figure 2 (a): Summary of sequence characteristics and activities of various candidate dsDNA cytosine deaminases homologous to DddAtox. #1: Sorted from highest to lowest according to PSI-BLAST scores. #2: +, indicating that the protein C-terminus contains one SPKK motif-like sequence; ++, indicating that the protein C-terminus contains two SPKK motif-like sequences. #3: *, indicating that the in vitro deamination activity is comparable to DddAtox; **, indicating that the in vitro deamination activity is higher than DddAtox; Weak, deamination products can be identified, but the average deamination rate is <5% at a protein concentration of 10 μM. Figure 2 (b): Probability map of mutated cytosine bases on the genome when Ddd_SS, Ddd_Fa and DddAtox are overexpressed in E. coli with UNG knockout.

[0003] Figure 3 The display shows the results of building a mitochondrial base editor using Ddd_SS. Figure 3 (a): Schematic diagram of DdCBE construction. Figure 3 (b): HEK293T cells transfected with ND5.1-DdCBE, mitochondrial DNA editing ratios after 3 days: A_G1397:DddAtox-G1397; SS_N94:Ddd_SS-N94; A_G1333:DddAtox-G1333; SS_N29:Ddd_SS-N29. The data shown in the figure are mean ± standard deviation, and the experiment was performed in 3 independent replicates.

[0004] Figure 4 This shows the editing function of the Ddd_SS mutant on the pathogenic gene. Figure 4 (a): Diseases caused by two types of mitochondrial gene mutations. Figure 4 (b): The proportion of mitochondrial DNA editing in HEK293T cells transfected with ND4-DdCBE 3 days later; Figure 4 (c): Mitochondrial DNA editing ratio in HEK293T cells 3 days after transfection with ND6-DdCBE; Figure 4 (b) and Figure 4In (c), the gray background indicates the disease-related sites in (a), and the gray bases are the TALE binding portions. Ddd_SS1:Ddd_SS(T26I+T77I+T110I); Ddd_SS2:Ddd_SS(T26I); Ddd_SS3:Ddd_SS(T77I); Ddd_SS4:Ddd_SS(T110I); Ddd_SS5:Ddd_SS(T77I+T110I); Dead-Ddd_SS5:Ddd_SS(E44A+T77I+T110I). Ddd_SS and its mutants use the N94 splitting site, and DddA and its mutants use the G1397 splitting site. The data shown in the figure are mean ± standard deviation, and the experiment was conducted in 3 independent replicates.

[0005] Figure 5 The DddAtox mutant expands the editable sites. Figure 5 (a): The proportion of mitochondrial DNA editing after 3 days of transfection of ND5.1-DdCBE into HEK293T cells; Figure 5 (b): Mitochondrial DNA editing ratio 3 days after transfection of ND1-DdCBE into HEK293T cells; Ddd_SS and its mutants used the N94 cleavage site, and DddA and its mutants used the G1397 cleavage site. The data shown in the figure are mean ± standard deviation, and the experiment was performed in 3 independent replicates. Detailed Implementation

[0006] The molecular biology methods used in this application, unless otherwise stated, are experimental methods well known to those skilled in the art. The reagents used, unless otherwise stated, are all commercially available and commonly used. Therefore, those skilled in the art can fully replicate the experiments of this application and obtain the same results by reading the following instructions.

[0007] Test methods

[0008] Strains and culture conditions

[0009] All strains were grown on Luria-Bertani (LB) medium or agar-solid LB medium at 37°C. Kanamycin (50 mg / L), ampicillin (100 mg / L), L-arabinose (2 g / L), and IPTG (0.5 mM) could be added to the medium as needed. Plasmids, protein expression, and heterologous expression of candidate deaminases were constructed and generated using *E. coli* DH5α, BL21(DE3), and BW25113Δung, respectively, to determine substrate preferences.

[0010] plasmid construction

[0011] To construct plasmids for deaminase expression, the deaminase gene and corresponding immunoprotein synthesized by Azenta Life Sciences (Suzhou, China) were cloned into MCS-1 (BamHI and NotI sites, introducing N-terminal hexahistine tags) and MCS-2 (NdeI and XhoI sites) of pCOLADuet1. To express deaminase in *E. coli* BW25113Δung, DddAtox, Ddd_SS, and Ddd_Fa were cloned downstream of the araBAD promoter in pBAD, and the corresponding immunoprotein gene driven by the T7 promoter was cloned downstream of the deaminase gene in the same plasmid. To assemble DdCBE, a TALE array was assembled from a tetrameric template using the Golden Gate cloning method. The TALE array was then digested with Ndel and BamH1 enzymes and ligated with cleavage deaminase, UGI, and other DdCBE sequences. The DdCBE construct in this study contained a mitochondrial targeting sequence from the TXN2 gene, a Flag / HA tag, a TALE array, cleavage deaminase, and UGI. Figure 2 Except for L-A1397N and L-SS94N in b and 2c, all DdCBEs were fused with GFP via self-cleavage of the T2A sequence. The DdCBEs were then cloned into the piggyBac vector and placed under the control of the CAGGS promoter.

[0012] Protein purification in in vitro DNA deamination analysis

[0013] To express deaminase, pCOLADuet-1 containing the genes for the deaminase and its corresponding immunoprotein was transformed into BL21(DE3). Single colonies were selected for characterization and amplification. Bacterial suspensions with an OD600 of 0.6–0.8 were incubated overnight at 18°C ​​with 0.5 mM IPTG to induce protein expression. Cells were harvested and resuspended in lysis buffer (50 mM Tris-HCl, pH 7.5, 500 mM NaCl, 5% glycerol, 20 mM imidazole, 5 mM 2-mercaptoethanol, and 1 mM PMSF) and lysed by sonication. The supernatant was separated by centrifugation at 18,000 rpm for 30 minutes in a JA-25.50 centrifuge (Beckman). The deaminase immunoprotein complex was purified from the cell lysate by nickel affinity chromatography using 1 mL Ni-Sepharoise 6 fast-flow agarose beads and loaded onto a gravity flow column (GE Healthcare). The supernatant was loaded onto the column and the resin was washed with 10 mL of wash buffer (50 mM Tris-HCl, pH 7.5, 500 mM NaCl, 20 mM imidazole, and 5 mM 2-mercaptoethanol). The deaminase-immunoprotein complex was eluted with 3 mL of elution buffer (50 mM Tris-HCl, pH 7.5, 300 mM imidazole, 500 mM NaCl, and 5 mM 2-mercaptoethanol), and the deaminase was then separated from the complex by denaturation and renaturation steps. For denaturation, the eluted protein sample was added to 25 mL of 6 M guanidine hydrochloride denaturation buffer (50 mM Tris-HCl, pH 7.5, 20 mM imidazole, 500 mM NaCl, and 5 mM 2-mercaptoethanol) and incubated at 4 °C for 1 h. The 6 M guanidine hydrochloride buffer containing the eluted protein was loaded onto a gravity flow column with 1 mL of Ni-Sepharoise 6 fast-flow agarose beads. The column was washed with 10 mL of 6M guanidine hydrochloride buffer to remove any remaining immunoproteins. While the deaminase was still bound to the Ni agarose beads, the protein was renatured by washing sequentially with 8 mL of denaturation buffer containing 10 μM ZnCl2 and decreasing concentrations of guanidine hydrochloride (5M, 4M, 3M, 2M, 1M), followed by washing with wash buffer to remove residual guanidine hydrochloride. The column-bound protein was then eluted with 3 mL of elution buffer. The eluted deaminase was further purified by size exclusion chromatography using a Superdex 75 column (GE Healthcare) in gel buffer (20 mM Tris-HCl pH 7.5, 200 mM NaCl, 5 mM 2-mercaptoethanol, and 5% glycerol). The fraction purity was assessed by SDS-PAGE after Coomassie blue staining, and the highest quality fraction was stored at -80°C.

[0014] DNA deamination test

[0015] The DNA deamination assay was performed largely as previously described, with some modifications. The DNA substrate was purchased from Sangon Biotech (Shanghai, China) and contained a 6-FAM fluorophore at its 5' end for visualization. To generate the double-stranded DNA substrate, unmodified reverse complementary oligonucleotides were annealed at equimolar concentrations to the 6-FAM fluorophore-modified substrate. The reaction was carried out in 10 μL of deamination buffer containing 20 mM Tris-HCl, 200 mM NaCl, 5 mM 2-mercaptoethanol, and 1 μM substrate at pH 7.5. The reaction was incubated at 37 °C for 1 hour, followed by the addition of 5 μL of UDG reaction solution (New England Biolabs, 1×UDG buffer containing 0.02 U / μL UDG) and incubation at 37 °C for another 30 minutes. UDG-mediated cleavage of the substrate at uracil residues was induced by the addition of 100 mM NaOH and incubation at 95 °C for 2 minutes. The reactants were analyzed by 20% acrylamide 8M urea gel electrophoresis in 1×TBE buffer, and the 6-FAM fluorophore signal was detected by fluorescence imaging using a ChemiDoc MP imaging system (Bio-Rad). The percentage of deamination was quantified using ImageJ.

[0016] Single nucleotide variant (SNV) analysis

[0017] To obtain genomic DNA, *E. coli* BW25113ung strain expressing candidate deaminases was inoculated at a 1:100 dilution into 20 mL of LB broth medium. The culture was grown to approximately OD600 of 0.6 and then treated with 2 g / L L-arabinose for 1 hour to induce deaminase expression. Using 3 mL of bacterial culture, bacterial genome was extracted using the FastPure Blood / Cell / Tissue / Bacterial DNA Isolation Mini Kit (Vazyme, DC112). Extraction yield was quantified using a Qubit (ThermoFisher Scientific). Sequencing libraries were constructed according to the manufacturer's instructions using the VAHTSUniversal Plus DNA Library Preparation Kit for Illumina V2 (Vazyme, ND627), except that the VAHTS HiFi Amplification Mix component was replaced with the KAPA HiFi HotStart Uracil+ReadyMix (KAPA Biosystems, KK2801) component to achieve efficient amplification of uracil encountered in the DNA template. Library concentration and quality were evaluated using Qubit and 1% agarose gel electrophoresis. Sequencing was performed using an Illumina Nova-seq 6000 sequencing system (Novogene), and readout maps of the reference genome (NC_000913.3) were plotted using BWA software (version 0.7.17). Duplications were removed using the Picard tool (version 2.18.29). Stacked data from alignments were generated using SAMtools (version 1.14), and variable calls were performed using VarScan (version 2.4.4). SNV validation thresholds were set to: variable frequency > 0.01, coverage > 50 reads per base, and p-value < 0.01. Probability logos for modifying consensus regions flanking bases were generated using the WebLogo online tool (https: / / weblogo.berkeley.edu).

[0018] Cell culture and transfection

[0019] HEK293T cells were cultured at 37°C and 5% CO2 in high-glucose DMEM medium (Hyclone, D6429) supplemented with 10% FBS (PANSera, 2602-P130707). To edit the mitochondrial genome, approximately 20 hours before transfection, 8 × 10⁶ cells were cultured... 4HEK293T cells were seeded in 24-well plates coated with poly-D-lysine 466 (PDL). JetTime transfection reagent (Polyplus-transfection, 468PT-114-75) was used, with 250 ng of each DdCBE monomer, for a total transfection volume of 500 ng of plasmid DNA. Cells were harvested 72 h post-transfection, and genomic DNA was extracted. Regions containing the target editing sites were then amplified by PCR for Sanger sequencing (Azenta Life Sciences) or library preparation for next-generation sequencing. EditR was used to assess the basic editing efficiency of the Sanger sequencing data.

[0020] Target amplicon sequencing and analysis

[0021] For target amplicon sequencing, the target region was first amplified by a first-round PCR using a KAPAHiFi HotStart Uracil+ReadyMix (KAPABiosystems, KK2801). Then, the Illumina index was increased by amplifying the first-round PCR product again in a second-round PCR using the N323 VAHTSRNAMultiplex Oligos Set 1 (Vazyme, N323) for Llumina. The PCR cycle number was optimized using qPCR to reach the top of the linear range to minimize amplification bias. For example, the inventors used 100 ng of whole-cell genome as the initial template, with 12 cycles in the first round of PCR and 10 cycles in the second round. The product was further purified using 1×VAHTSDNA Clean Beads (Vazyme, N411), and the library was then sequenced using an Illumina Nova seq 6000 sequencing system. Target amplicon sequencing analysis was performed using CRISPResso2. The output file "Nucleode_percentage_table.txt" is used to edit the frequency quantization.

[0022] statistics

[0023] Unless otherwise stated, data are presented as mean ± standard deviation, and statistical analysis was performed using GraphPad Prism 9.

[0024] Example 1: Modification of DddAtox

[0025] From the amino acid sequence analysis of DddAtox, the inventors noted that its C-terminus contains two SPKK-related peptide motifs. Figure 1(a) These motifs are known to preferentially bind to A / T-rich DNA sequences located in the minor groove of double-stranded DNA. Deletion of these two SPKK-related motifs completely eliminates the double-stranded DNA deaminase activity of DddAtox, while adding an AT-hook with similar DNA-binding properties to the SPKK-related motifs restores the deaminase activity of the truncated DddAtox. Figure 1 (b) and Figure 1 (c)). These data indicate that the SPKK-related motif at the C-terminus of DddAtox is important for the deamination activity of its double-stranded DNA.

[0026] Next, the inventors used PSI-BLAST from the MPI Bioinformatics Toolkit to search for homologs of DddAtox. They ran a non-redundant (NR) protein database (2021, nr50_1_Nov) iteratively until no new sequences appeared, ultimately identifying 555 homologous candidate sequences. From these, they selected 8 candidate proteins: 4 with SPKK-related motifs and 4 without, and tested their deaminase activity on double-stranded DNA substrates. Figure 2 (a)). By Figure 2 (a) As can be seen, all four proteins containing the SPKK-related motif exhibited deamination activity comparable to or higher than that of DddAtox. Conversely, all four proteins lacking the SPKK-related motif showed little or no deamination activity. These results suggest that the SPKK-related motif can be used to identify highly active double-stranded DNA deaminases.

[0027] The sequences of four proteins with SPKK-related motifs are as follows:

[0028] 1. Ddd_SS, derived from Simiaoa Sunii:

[0029] MSLPEYDGTTTHGVLVLDDGTQIGFTSGNGDPRYTNYRNNGHVEQKSALYMRENNISNATVYHNNTNGTCGYCNTMTATFLPEGATLTVVPPENAVANNSRAIDYVKTYTGTSNDPKISPRYKGN (SEQ ID NO: 1)

[0030] 2. Ddd_Ru, derived from Ruminococcus sp. MSJ-25:

[0031] VLPKYDGKTTEGVMVTPDGKQISFKSGNSSTPSYPQYKAQSASHVEGKAALYMRENGINEATVFHNNPNGTCGFCDRQVPALLPKGAKLTVVPPSSNSVANNVRAIPVPKTYIGNSTVPKIK(SEQ ID NO: 2)

[0032] 3. Ddd_Fa, derived from Fusarium MSJ-15 (Falcatimonas sp. MSJ-15):

[0033] SINLPEYDGKTTHGVLVLDDGTQVPFSSGNANPNYKNYIPASHVEGKSAIYMRENGINNGTVFHNNDTGTCPYCDKMLPTLLEEGSTLTVVPPANANAPKPSWVDTVKTYIGNDKIPKKPK(SEQ ID NO: 3)

[0034] 4. Ddd_CA, derived from *Chondromyces apiculatus*:

[0035] MGNTLPGWDGGKTQGWFVYPDGTERHLISGYDGPSKFTQGIPGMNGNIKSHVEAHAAALMRQYELSKATLYINRVPCPGVRGCDALLARMLPEGVQLEIIGPNGFKKTYTGLPDPKLKPKGCS(SEQ ID NO: 4)

[0036] In deamination experiments, the inventors observed multiple bands formed by the double-stranded DNA deaminase (Ddd_SS) from *Simiaoa Sunii*, indicating its broad deaminase activity in non-TC environments. In fact, whole-genome sequencing of a UNG-deficient *E. coli* strain expressing Ddd_SS showed a preference for deamination activity at AC / GC / TC sequence sites, while another double-stranded DNA deaminase, Ddd_Fa, exhibited deamination activity at AC / CC / TC sequence sites with a slightly higher preference for TC, and the original DddAtox showed a very high preference for TC sequence sites. Figure 2 (b) Since Ddd_SS possesses deamination activity at GC sequence sites that DddAtox and its derivatives do not have, and exhibits the highest deamination activity among all tested deaminases (including DddAtox) (comparing the product yield of deaminases at 0.5 μM), the inventors focused on Ddd_SS in subsequent studies.

[0037] Example 2: Base editing effect of the Ddd_SS construct on mitochondrial DNA

[0038] In this embodiment, the inventors tested the gene editing capabilities of Ddd_SS in target mitochondrial DNA by fusing half of Ddd_SS into a TALE array protein containing a mitochondrial localization sequence (MTS). Based on the alignment of the Ddd_SS structure predicted by ColabFold with the crystal structure of DddAtox, N29 and N94 in Ddd_SS correspond to the optimal splitting sites G1333 and G1397 reported by DddAtox6. Based on the initial DdCBE studies, the inventors designed the following mitochondrial DdCBE containing Ddd_SS (DdCBE_SS): it is the product of linking a pair of mitochondrial TALEs with a cleaved Ddd_SS, containing an MTS, a TALE array, half of the Ddd_SS cleaved from the N29 or N94 site, and a UGI protein. The C-terminal and N-terminal halves of Ddd_SS were respectively linked to the right and left sides of the TALEs (…). Figure 3 (a)).

[0039] The inventors first tested DdCBE_SS targeting MT-ND5, which encodes the NADH dehydrogenase 5 subunit of complex I. The inventors found that DdCBE_SS can achieve C→T editing at poly-C sites and exhibits different sequence preferences at the same level compared to DdCBEs containing DddAtox. Importantly, DdCBE_SS with the N94 cleavage site (DdCBE_SS_N94) shows a higher sequence preference at the G... C Editing efficiency for C6 at the site is approximately 40%, while DdCBE_A (DdCBE_A_G1397) with the G1397 cleavage site has an editing efficiency of only about 8% for C6. Furthermore, DdCBE_SS_N94 at G... C Editing efficiency at site C7 was approximately 33 times higher than that of DdCBE_A_G1397. For this MT-ND5.1 site, the DdCBE_SS_N29 construct showed lower editing efficiency at both GC sites than the DdCBE_SS_N94 construct (6.1% for C6 and 5.2% for C7), but this was still significantly higher than the DdCBE_A_G1333 construct (less than 0.3% at both sites). Figure 3 (b)). These results indicate that the inventors successfully modified DdCBE using Ddd_SS, achieving the desired performance in G. C At the site compared to DddA tox A much more efficient mtDNA editing method. It should be noted that because base editors can only edit one strand of dsDNA, the theoretically highest editing efficiency measured using the above method is 50%.

[0040] To test the universality of DdCBE_SS in mitochondrial DNA, the inventors subsequently constructed DdCBE_SS targeting MT-ATP6. The inventors observed that DdCBE_SS_N94 achieved the highest editing efficiency of 38%, and DdCBE_SS_N29 achieved the highest editing efficiency of 24%. Figure 2 c and Figure 3 Encouraged by these results, the inventors subsequently tested editing at eight additional sites in six mitochondrial genes (Table 2), where DdCBE_SS_N94 achieved editing efficiencies of approximately 7-42%. Figure 4 (a-4c). In summary, these results indicate that DdCBE_SS is a versatile and efficient mitochondrial DNA editing tool.

[0041] Example 3: Editing function of DdCBE_SS on pathogenic mutant genes

[0042] In this embodiment, the inventors introduced two mutations at the GC sites of MT-ND4 and MT-ND6, two genes associated with human diseases Leber hereditary optic neuropathy (LHON) and Leigh syndrome. Figure 2 d) Related, both diseases are devastating genetic disorders with no effective treatment currently available. DdCBE_SS_N94 produces a 10% editing efficiency on MT-ND4(C8). Figure 4 (b) shows approximately 20% improved editing efficiency for MT-ND6(C8). Figure 4 (c) To further improve the efficiency of DdCBE_SS_N94, the inventors further mutated it, altering the T26I, T77I, and T110I sites (corresponding to S1330I, T1380I, and T1413I of DddAtox, respectively). Excitingly, the DdCBE_SS_N94 mutant (DdCBE_SS5) with the T77I and T110I mutations showed significantly improved editing efficiency at disease-related target sites, achieving an editing efficiency of approximately 25% for MT-ND4(C8). Figure 4 (b) The editing efficiency for MT-ND6(C8) is approximately 30%. Figure 4 (c)). These results demonstrate that the optimized DdCBE obtained by the inventors from Ddd_SS can be used in G C The site enables disease-related mtDNA mutations, which was previously impossible.

[0043] The amino acid sequences of Ddd_SS containing each mutation are shown below:

[0044] Ddd_SS1:Ddd_SS(T26I+T77I+T110I);

[0045] SLPEYDGTTTHGVLVLDDGTQIGF I SGNGDPRYTNYRNNGHVEQKSALYMRENNISNATVYHNNTNGTCGYCNTM I ATFLPEGATLTVVPPENAVANNSRAIDYVKTY I GTSNDPKISPRYKGN(SEQ ID NO:5)

[0046] Ddd_SS2:Ddd_SS(T26I);

[0047] SLPEYDGTTTHGVLVLDDGTQIGF I SGNGDPRYTNYRNNGHVEQKSALYMRENNISNATVYHNNTNGTCGYCNTMTATFLPEGATLTVVPPENAVANNSRAIDYVKTYTGTSNDPKISPRYKGN(SEQ ID NO:6)

[0048] Ddd_SS3:Ddd_SS(T77I);

[0049] SLPEYDGTTTHGVLVLDDGTQIGFTSGNGDPRYTNYRNNGHVEQKSALYMRENNISNATVYHNNTNGTCGYCNTM I ATFLPEGATLTVVPPENAVANNSRAIDYVKTYTGTSNDPKISPRYKGN(SEQ ID NO:7)

[0050] Ddd_SS4:Ddd_SS(T110I);

[0051] SLPEYDGTTTHGVLVLDDGTQIGFTSGNGDPRYTNYRNNGHVEQKSALYMRENNISNATVYHNNTNGTCGYCNTMTATFLPEGATLTVVPPENAVANNSRAIDYVKTY I GTSNDPKISPRYKGN(SEQ ID NO:8)

[0052] Ddd_SS5:Ddd_SS(T77I+T110I);

[0053] SLPEYDGTTTHGVLVLDDGTQIGFTSGNGDPRYTNYRNNGHVEQKSALYMRENNISNATVYHNNTNGTCGYCNTM I ATFLPEGATLTVVPPENAVANNSRAIDYVKTY I GTSNDPKISPRYKGN(SEQ ID NO:9)

[0054] Dead-Ddd_SS5:Ddd_SS(E44A+T77I+T110I),

[0055] SLPEYDGTTTHGVLVLDDGTQIGFTSGNGDPRYTNYRNNGHV A QKSALYMRENNISNATVYHNNNTGTCGYCNTM I ATFLPEGATLTVVPPENAVANNSRAIDYVKTY I GTSNDPKISPRYKGN(SEQ ID NO:10)

[0056] Example 4: DddAtox mutant expands editable sites

[0057] DdCBE_SS exhibits a different sequence preference than DdCBE_A. Previous studies on the C→U RNA editing enzyme APOBEC have shown that loop sequences are important for the sequence preference of enzyme activity. To identify the differential sequences determining the different sequence preferences of Ddd_SS and DddAtox, the inventors mutated three loop sequences near the active site of DddAtox, based on the loop sequence of Ddd_SS. Using the MT-ND5.1 site as the target, the DdCBE_A mutant with the loop 2 mutation (E1370N) showed an approximately 3.2-fold increase in editing efficiency at C6 (GC site), an approximately 2.0-fold increase at C13 (AC site), and an approximately 2.3-fold increase at C14 (CC site). Figure 5 b). In contrast, the ring 3 mutation has almost no effect on the activity of DdCBE_A, while the ring 1 mutation slightly decreases the activity of DdCBE_A. Figure 5 (a) Furthermore, compared to the original DdCBE_A, the DdCBE_A mutant with the ring 2 mutation (E1370N) also achieved higher editing efficiency at the MT-ND1 and MT-ND5.2 sites. Figure 5 (b)). In summary, these results demonstrate that sequence compatibility and editing efficiency of DDCBE can be reasonably optimized by exchanging sequences from different homologs.

[0058] As can be seen from the above embodiments, the inventors of this invention discovered that the SPKK-related motif at the C-terminus of proteins is crucial for the DNA deamination efficiency of DddAtox. Furthermore, the inventors identified numerous DddAtox homolog candidates using PSI-BLAST and confirmed four homologs with double-stranded DNA deaminase activity, all of which possess SPKK-related motifs. Then, the inventors constructed multiple DDCBEs from Ddd_SS, capable of efficiently editing 14 mitochondrial DNA sites in 10 mitochondrial genes. Importantly, the DdCBE_SS variant successfully achieved previously unattainable C→T efficient editing at GC sites in mitochondrial DNA. Finally, by introducing the Ddd_SS mutant into DddAtox, the inventors successfully constructed a DdCBE_A with broader sequence compatibility and higher editing efficiency. In summary, this application provides previously unattainable C→T editing at GC sites in mitochondrial DNA and opens up possibilities for further screening and designing mtDNA base editors with potentially higher efficiency and broader sequence compatibility.

Claims

1. A kind of enzyme with the deamination activity of cytosine on GC sequence site of double-stranded DNA in mitochondria, which is a protein obtained by point mutation of T77I and / or T110I to the amino acid sequence shown in SEQ ID NO:

1. 2.A base editing system for deamination of cytosine on GC sequence site of double-stranded DNA in mitochondria, comprising N-terminal half and C-terminal half of the enzyme as claimed in claim 1, which is split from N94 site or N29 site and connected with a pair of transcription activator-like effectors (TALE) respectively. 3.The base editing system of claim 2, further comprising uracil DNA glycosylase inhibitor (UGI) protein.

Images