Compositions and methods for epigenetic regulation of PCSK9 expression

Abstract

This application relates to compositions and methods comprising an epigenetic editor for the epigenetic modification of PCSK9, as well as nucleic acids and vectors encoding the same. It also discloses cells that are epigenetically modified by the epigenetic editor.

Description

Technical Field

[0001] Cross - reference to Related Applications This application claims the benefit of U.S. Provisional Patent Application No. 63 / 488,736, filed on March 6, 2023, entitled "COMPOSITIONS AND METHODS FOR EPIGENETIC REGULATION OF PCSK9 EXPRESSION" under 35 U.S.C. § 119(e), the entire disclosure of which is incorporated herein by reference in its entirety.

[0002] Reference to Electronic Sequence Listing The content of the electronic sequence listing (C169870005WO01 - SEQ - AXW.xml, size: 1,866,473 bytes, and creation date: March 6, 2024) is incorporated herein by reference in its entirety.

Background Art

[0003] Genome editing has been considered a promising therapeutic approach for treating genetic diseases for over a decade. However, DNA - level manipulation using conventional gene editors still poses risks considering unwanted double - strand DNA breaks, non - uniform repair (including insertions and deletions of various sizes at the intended site), and the potential for toxicity. In contrast, targeted epigenetic modifications offer the ability to alter gene expression without inducing the genotoxicity associated with double - strand breaks.

[0004] One promising candidate for epigenetic silencing is the proprotein convertase subtilisin / kexin type 9 (PCSK9) gene. PCSK9 is a key target in the treatment of heart disease, a leading cause of death worldwide (Berberich et al., Nature Rev Cardiol. (2019) 16(1):9-20). The human PCSK9 gene, located on chromosome 1, shares approximately 94% and 80% homology with its cynomolgus monkey and mouse counterparts, respectively. This gene has a CpG island in its promoter region, which is distal to other genes and cis-regulatory properties. The PCSK9 protein is primarily produced by the liver.

[0005] In humans, PCSK9 plays a crucial role in regulating the circulating levels of low-density lipoprotein (LDL) particles as a result of binding to the LDL receptor (LDLR). LDLR reduces the circulating concentration of LDL particles by mediating endocytosis and degradation of LDL particles in cells. In the absence of PCSK9, or when the interaction between PCSK9 and LDLR is blocked, the recycling rate of LDLR to the cell surface increases, and recycled LDLR proteins continue to remove LDL particles from the extracellular fluid (Tombling et al., Atherosclerosis (2021) 330:52-60). In contrast, when endocytized LDLR is bound to PCSK9, LDLR is degraded along with its passenger LDL particles. Clinical and genetic studies have established that circulating LDL contributes to atherosclerotic cardiovascular disease (Ference et al., Eur Heart J (2017) 38:2459-72). In addition, loss-of-function mutations in PCSK9 are associated with low LDL levels (Zhao et al., Am J Hum Genet. (2006) 79(3):514-23). ​​Genetic or pharmacological reduction of PCSK9 reduces cardiovascular events (Ference et al., N Engl J Med. (2016) 375(22):2144-53, Sabatine et al., N Engl J Med. (2017) 376(18):1713-22). Reducing PCSK9 expression may help increase LDLR recycling, which would result in lower blood LDL particle concentrations. [Overview of the project] [Problems that the invention aims to solve]

[0006] Given the crucial role of PCSK9 in the pathogenesis of hypercholesterolemia and cardiovascular disease, there is a need for novel and improved therapies that target PCSK9 expression.

[0007] This disclosure provides systems and compositions for epigenetic modification (hereinafter referred to as "epigenetic editors" or "epigenetic editing systems"), and methods for using them to generate epigenetic modifications to PCSK9 in, for example, host cells and organisms. [Means for solving the problem]

[0008] In some embodiments, the present disclosure relates to a system for repressing the transcription of the human PCSK9 gene in human cells, and possibly human hepatocytes. a) A domain of DNA methyltransferase (DNMT) and / or a domain that recruits DNMT, wherein the DNMT domain and / or recruiter domain may include a DNMT3A domain and / or a DNMT3L domain, and the recruited DNMT may be DNMT3A, and Transcriptional repressor domain It collectively includes, A fusion protein in which each domain is linked to a DNA-binding domain that binds to a target region in the human PCSK9 gene, or b) One or more nucleic acid molecules encoding one or more fusion proteins We provide a system that includes this.

[0009] In some embodiments, the DNA-binding domain binds to the target sequence in SEQ ID NO: 1488 or 1489. In a particular embodiment, the DNA-binding domain targets one or more sequences in the PCSK9 gene selected from SEQ ID NOs: 700-747 and 1036-1261.

[0010] In some embodiments, the DNA-binding domain includes a dead CRISPR-Cas (dCas) domain, a ZFP domain, or a TALE domain. For example, the DNA-binding domain may include a dCas9 domain, and the system may further include (i) one or more guide RNAs (e.g., one of sequence numbers 1262-1487), or (ii) a nucleic acid molecule encoding one or more guide RNAs. In a particular embodiment, the dCas domain includes a sequence having at least 90% identity to a dCas9 sequence, e.g., sequence number 12 or 13.

[0011] In some embodiments, the fusion protein comprises a dead CRISPR-Cas (dCas) domain, and the system comprises one or more PCSK9-binding guide RNAs (gRNAs) provided herein. In some embodiments, the system comprises a single gRNA. In some embodiments, the system comprises two gRNAs. In some embodiments, the system comprises three gRNAs. In some embodiments, the system comprises four gRNAs. In some embodiments, the system comprises five or more gRNAs. In some embodiments, the system comprises gRNAs selected from those provided in Table 2. In some embodiments, the system comprises gRNAs selected from those provided in Table 7. In some embodiments, the system comprises gRNAs selected from those provided in Table 8. In some embodiments, the system comprises sgRNAs selected from those provided in Table 10. In some embodiments, the system comprises gRNAs selected from those provided in Table 12.

[0012] In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA009 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA009. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA003 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA003. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA093 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA093. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA011 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA011. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA007 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA007. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA077 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA077. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA113 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA113. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA004 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA004. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA008 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA008. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA012 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA012. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA111 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA111. In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA005 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA005.In some embodiments, the system includes a gRNA containing the gRNA targeting sequence of gRNA013 in Table 10, or a gRNA that binds to the same target domain sequence as gRNA013.

[0013] In some embodiments, the system includes gRNA g041 provided in Table 12, or a gRNA that binds to the target domain sequence of g041. In some embodiments, the system includes gRNA g049 provided in Table 12, or a gRNA that binds to the target domain sequence of g049. In some embodiments, the system includes gRNA g056 provided in Table 12, or a gRNA that binds to the target domain sequence of g056. In some embodiments, the fusion protein includes the fusion protein disclosed in Example 12. In some embodiments, the system includes the fusion protein provided in Example 12 and the gRNA provided in Table 10. In some embodiments, the system includes the fusion protein provided in Example 12 and the gRNA provided in Table 12.

[0014] In some embodiments, the system includes fusion protein 9 variant 1 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 9 variant 2 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 10 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 11 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 12 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 13 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 14 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 15 (Example 12) and gRNA g041. In some embodiments, the system includes fusion protein 9 variant 1 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 9 variant 2 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 10 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 11 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 12 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 13 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 14 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 15 (Example 12) and gRNA g049. In some embodiments, the system includes fusion protein 9 variant 1 (Example 12), as well as gRNA g041 and g049. In some embodiments, the system includes fusion protein 9 variant 2 (Example 12), as well as gRNA g041 and g049.In some embodiments, the system includes fusion protein 10 (Example 12), and gRNAs g041 and g049. In some embodiments, the system includes fusion protein 11 (Example 12), and gRNAs g041 and g049. In some embodiments, the system includes fusion protein 12 (Example 12), and gRNAs g041 and g049. In some embodiments, the system includes fusion protein 13 (Example 12), and gRNAs g041 and g049. In some embodiments, the system includes fusion protein 14 (Example 12), and gRNAs g041 and g049. In some embodiments, the system includes fusion protein 15 (Example 12), and gRNAs g041 and g049.

[0015] In some embodiments, the DNA-binding domain includes a ZFP domain that targets a nucleotide sequence selected from SEQ ID NOs. 700–747. In a particular embodiment, the ZFP domain sequentially includes one of the F1–F6 amino acid sequences from ZF001–ZF048 shown in Table 1.

[0016] In some embodiments, the DNMT3A domain includes a sequence having at least 90% identity with SEQ ID NO: 574 or 575.

[0017] The DNMT3L domain may contain a sequence having at least 90% identity to a sequence selected from, for example, sequence numbers 578-581. In some embodiments, the DNMT3L domain contains a sequence having at least 90% identity to a sequence selected from sequence numbers 582-603. In some embodiments, the DNMT3L domain contains a sequence having at least 90% identity to a sequence selected from sequence numbers 601-603.

[0018] In some embodiments, the transcriptional repressor domain includes a sequence having at least 90% identity to a sequence selected from SEQ ID NOs: 33-570. In certain embodiments, the transcriptional repressor domain is a KRAB domain derived from KOX1, ZIM3, ZFP28, or ZN627. The KRAB domain may include a sequence having at least 90% identity to a sequence selected from, for example, SEQ ID NOs: 89, 116, 245, and 255. In some embodiments, the transcriptional repressor domain includes a fusion of the N-terminal and C-terminal regions of ZIM3 and KOX1 KRAB, and optionally includes the amino acid sequence of SEQ ID NOs: 571 or 572. In certain embodiments, the transcriptional repressor domain is derived from KAP1, MECP2, HP1a / CBX5, HP1b, CBX8, CDYL2, TOX, TOX3, TOX4, EED, EZH2, RBBP4, RCOR1, or SCML2.

[0019] In some embodiments, the system a) comprising a DNMT3A domain, a DNMT3L domain, a transcriptional repressor domain, and a DNA-binding domain, In some cases, one or both of the DNMT3A and DNMT3L domains are human. In some cases, the DNA-binding domain is a dead CRISPR-Cas domain or a ZFP domain. Fusion protein, or b) Nucleic acid molecules encoding a fusion protein Includes.

[0020] In certain embodiments, the fusion protein comprises, from the N-terminus to the C-terminus, a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a DNA binding domain, a third peptide linker, and a transcriptional repressor domain. For example, the fusion protein can comprise, from the N-terminus to the C-terminus, a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a first nuclear localization signal (NLS), a DNA binding domain, a second NLS, a third peptide linker, and a transcriptional repressor domain. The fusion protein can comprise, from the N-terminus to the C-terminus, a first NLS, a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a DNA binding domain, a third peptide linker, a transcriptional repressor domain, and a second NLS. The fusion protein can comprise, from the N-terminus to the C-terminus, first and second NLSs, a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a DNA binding domain, a third peptide linker, a transcriptional repressor domain, and third and fourth NLSs. In certain embodiments, the transcriptional repressor domain is a KRAB domain, such as the human KOX1, ZFP28, ZN627, or ZIM3 KRAB domain. In certain embodiments, one or both of the second and third peptide linkers are XTEN linkers, which can be selected from XTEN80 (e.g., SEQ ID NO: 643) and XTEN16 (e.g., SEQ ID NO: 638), for example, the second peptide linker is XTEN80 and the third peptide linker is XTEN16.

[0021] In some embodiments, the fusion protein can comprise, from the N-terminus to the C-terminus, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a first NLS, a dSpCas9 domain, a second NLS, an XTEN16 peptide linker, and a human KOX1 KRAB domain. In certain embodiments, the fusion protein comprises a sequence that is at least 90% identical to SEQ ID NO: 658.

[0022] In some embodiments, the fusion protein comprises, from the N-terminus to the C-terminus, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a first NLS, a ZFP domain, a second NLS, an XTEN16 linker, and a human KOX1 KRAB domain. In certain embodiments, the fusion protein comprises a sequence that is at least 90% identical to SEQ ID NO: 659, and optionally, the ZFP comprises any one of the F1-F6 amino acid sequences of ZF001-ZF048 shown in Table 1 in sequence.

[0023] In some embodiments, the fusion protein comprises, from the N-terminus to the C-terminus, a first and a second NLS, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a dSpCas9 domain, an XTEN16 peptide linker, a human KOX1 KRAB domain, and a third and a fourth NLS. In certain embodiments, the fusion protein may comprise the amino acid sequence of SEQ ID NO: 660 or a sequence that is at least 90% identical thereto.

[0024] In some embodiments, the fusion protein comprises, from the N-terminus to the C-terminus, a first and a second NLS, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a ZFP domain, an XTEN16 peptide linker, a human KOX1 KRAB domain, and a third and a fourth NLS.

[0025] In some embodiments, the fusion protein comprises, from the N-terminus to the C-terminus, a first and a second NLS, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a dSpCas9 domain, an XTEN16 peptide linker, a human ZFP28 KRAB domain, and a third and a fourth NLS. In certain embodiments, the fusion protein may comprise the amino acid sequence of SEQ ID NO: 661 or a sequence that is at least 90% identical thereto.

[0026] In some embodiments, the fusion protein comprises, from N-terminus to C-terminus, first and second NLSs, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a ZFP domain, an XTEN16 peptide linker, a human ZFP28 KRAB domain, and third and fourth NLSs.

[0027] In some embodiments, the fusion protein comprises, from N-terminus to C-terminus, first and second NLSs, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a dSpCas9 domain, an XTEN16 peptide linker, a human ZN627 KRAB domain, and third and fourth NLSs. In certain embodiments, the fusion protein may contain the amino acid sequence of SEQ ID NO: 662 or a sequence that is at least 90% identical thereto.

[0028] In some embodiments, the fusion protein comprises, from N-terminus to C-terminus, first and second NLSs, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a ZFP domain, an XTEN16 peptide linker, a human ZN627 KRAB domain, and third and fourth NLSs.

[0029] In some embodiments, the fusion protein comprises, from N-terminus to C-terminus, first and second NLSs, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a dSpCas9 domain, an XTEN16 peptide linker, a human ZIM3 KRAB domain, and third and fourth NLSs. In certain embodiments, the fusion protein may contain the amino acid sequence of SEQ ID NO: 663 or a sequence that is at least 90% identical thereto.

[0030] In some embodiments, the fusion protein comprises, from N-terminus to C-terminus, first and second NLSs, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a ZFP domain, an XTEN16 peptide linker, a human ZIM3 KRAB domain, and third and fourth NLSs.

[0031] In some embodiments, at least one of the NLSs in the fusion protein described herein is SV40 NLS (e.g., SEQ ID NO: 644).

[0032] In some embodiments, the system a) A first fusion protein comprising a first DNA-binding domain and comprising or mobilizing a DNMT3A domain, A second fusion protein containing a second DNA-binding domain and containing or mobilizing a DNMT3L domain, A third fusion protein comprising a third DNA-binding domain and comprising or mobilizing a transcriptional repressor domain, or b) One or more nucleic acid molecules encoding a fusion protein Includes.

[0033] This disclosure also provides human cells, or offspring of such cells, comprising the system described herein. In some embodiments, the cells are hepatocytes.

[0034] This disclosure also provides pharmaceutical compositions comprising the systems described herein and pharmaceutically acceptable excipients.

[0035] This disclosure also provides a method for treating a patient in need of treatment, comprising the step of administering to the patient (e.g., intravenously) a system or pharmaceutical composition described herein. In some embodiments, the patient has heart disease; elevated or hypercholesterolemia of low-density lipoprotein cholesterol (LDL-C); at risk of developing myocardial infarction, stroke, or unstable angina; and / or primary hyperlipidemia (e.g., heterozygous familial hypercholesterolemia (HeFH) or homozygous familial hypercholesterolemia (HoFH)).

[0036] This disclosure also provides systems or pharmaceutical compositions described herein for use in, for example, the methods described herein, in treating patients in need of treatment.

[0037] This disclosure also provides the use of the system described herein in the manufacture of a pharmaceutical product for treating patients in need of treatment, for example, in the manner described herein.

[0038] This disclosure also provides articles and kits including the systems described herein.

[0039] Other properties, purposes, and advantages of the disclosed methods and compositions will become apparent in the following detailed description. However, it should be understood that the detailed description is provided for illustrative purposes only, not limiting purposes, while illustrating embodiments and examples of the disclosed methods and compositions. Various changes and modifications within the scope of this disclosure will be apparent to those skilled in the art from the detailed description. [Brief explanation of the drawing]

[0040] [Figure 1] This figure shows the ZF protein on the PCSK9 gene and the predicted binding sites of the computationally designed gRNA. [Figure 2]This scatter plot shows the relative PCSK9 expression (y axis) in cells treated with CRISPR-off (DNMT3A-3L-dCas9-KRAB) on day 7. The genomic distance from the gRNA target site to the PCSK9 TSS is shown on the x axis. [Figure 3] This figure shows the duplication of the top 40 gRNAs with the PCSK9 gene. [Figure 4A] This bar graph shows the levels of secreted PCSK9 at 7 and 28 days after treatment with the indicated gRNA. The dashed line indicates silencing achieved by wild-type (WT) Cas9. [Figure 4B] This is a scatter plot showing the correlation between PCSK9 mRNA expression and PCSK9 protein secretion in cells treated with gRNA. CRISPRi(dCas9-KRAB) represents the dCas9-KRAB fusion protein. [Figure 5] This line graph shows the silencing of PCSK9 after treatment with CRISPRi (dCas9-KRAB), CRISPR-off (DNMT3A-3L-dCas9-KRAB), and the indicated gRNAs. [Figure 6] This bar graph shows PCSK9 secretion in cells treated with CRISPRoff and simvastatin compared to cells treated with the CRISPRoff system alone. [Figure 7] This bar graph shows the reduction in PCSK9 secretion in Huh7 hepatoma cells treated with CRISPRoff and a given gRNA. [Figure 8] This scatter plot shows the activity and toxicity of 247 PCSK9-targeted ZF proteins. Relative PCSK9 expression is shown on the x-axis, and the corresponding cell count compared to pUC and off-target controls is shown on the y-axis. The diagonal line represents the 1:1 correlation between relative PCSK9 expression and cell count. [Figure 9] This scatter plot shows the relative PCSK9 expression (y-axis) and the corresponding targeted genomic distance (x-axis) to the PCSK9 transcription start site (TSS) in cells treated with the ZF-off (DNMT3A-3L-ZF-KRAB) construct. [Figure 10] This figure shows the entire human PCSK9 locus adjacent to 35.5kb and 7kb of the upstream and downstream genomic regions (67.5kb), respectively, as introduced into and expressed in transgenic mice. The transgenic mouse line expresses human PCSK9 under the control of its own (human) endogenous promoter. [Figure 11] Figure 11A shows a schematic diagram of a fusion protein construct having the variant NLS configuration. Figure 11B shows a schematic diagram of an additional fusion protein construct having the variant KRAB domain. [Figure 12A] Figures 12A and 12B are graphs showing the percentage of PCSK9 protein levels measured after treatment with fusion protein constructs having various NLS configurations in HeLa cells, using 6.25 ng of RNA (Figure 12A) or 2.5 ng of RNA (Figure 12B). Human and mouse DNMT3L sequences are shown as h3L and m3L, respectively. [Figure 12B] Figures 12A and 12B are graphs showing the percentage of PCSK9 protein levels measured after treatment with fusion protein constructs having various NLS configurations in HeLa cells, using 6.25 ng of RNA (Figure 12A) or 2.5 ng of RNA (Figure 12B). Human and mouse DNMT3L sequences are shown as h3L and m3L, respectively. [Figure 13] This graph shows that constructs containing 2X NLS are three times more efficient than CRISPR-off in silencing mPcsk9 in Hepa1-6 cells. [Figure 14A] Figure 14A is a graph showing that constructs with 2X NLS are more efficient than CRISPR-off in silencing mPcsk9 in Huh7 cells. [Figure 14B]Figures 14B–14C are graphs showing that constructs with 2X NLS are also more efficient than CRISPR-off in silencing mPcsk9 in Huh7 cells at different doses at both day 5 (Figure 14B) and day 15 (Figure 14C). [Figure 14C] Figures 14B–14C are graphs showing that constructs with 2X NLS are also more efficient than CRISPR-off in silencing mPcsk9 in Huh7 cells at different doses at both day 5 (Figure 14B) and day 15 (Figure 14C). [Figure 15] This graph shows that in Huh7 cells, 2X NLS provides improvement across multiple ZFs in a CRISPR-off-like configuration where dCas9 is replaced by a zinc finger. [Figure 16] This is a set of graphs showing that methylation of the CTLA4 promoter in bacterial DNMT proteins can induce epigenetic silencing of the gene locus. [Figure 17] This is a set of graphs showing the methylation profiles at the VIM3 locus of cells treated with different constructs containing dCas9-fused bacterial DNA methyltransferase at day 30. Samples treated with M. SssI are 20% methylated. [Figure 18] This is a set of graphs showing the methylation profiles of the cellular CLTA locus, comparing M. SssI with mouse DNMT3A / 3L in dCas9 fusions on day 29, mediated by hybridization capture. [Figure 19A] This is a set of graphs showing alternative KRAB domains tested for episilencing activity against CRISPR-off when CLTA-GFP is used as a marker and 0.5 ng of effector DNA is used (Figure 19A), when GFP is used as a marker and 3 ng of effector DNA is used (Figure 19B), and when GFP is used as a marker and 0.5 ng of effector DNA is used (Figure 19C). [Figure 19B] This is a set of graphs showing alternative KRAB domains tested for episilencing activity against CRISPR-off when CLTA-GFP is used as a marker and 0.5 ng of effector DNA is used (Figure 19A), when GFP is used as a marker and 3 ng of effector DNA is used (Figure 19B), and when GFP is used as a marker and 0.5 ng of effector DNA is used (Figure 19C). [Figure 19C] This is a set of graphs showing alternative KRAB domains tested for episilencing activity against CRISPR-off when CLTA-GFP is used as a marker and 0.5 ng of effector DNA is used (Figure 19A), when GFP is used as a marker and 3 ng of effector DNA is used (Figure 19B), and when GFP is used as a marker and 0.5 ng of effector DNA is used (Figure 19C). [Figure 19D] Figure 19D shows the results after 30 days using various nanogram amounts of effector DNA. [Figure 20A] Figures 20A and 20B show the experimental timeline (Figure 20A) and experimental design (Figure 20B) used to test PCSK9 silencing in transgenic mice expressing human PCSK9 (hPCSK9), as described in Example 11. [Figure 20B] Figures 20A and 20B show the experimental timeline (Figure 20A) and experimental design (Figure 20B) used to test PCSK9 silencing in transgenic mice expressing human PCSK9 (hPCSK9), as described in Example 11. [Figure 21A] Figures 21A and 21B are a set of graphs showing the silencing of PCSK9 in transgenic mice expressing human PCSK9, measured over 42 days (Figure 21A) and 84 days (Figure 21B). As described in Example 11, human PCSK9 levels in the mouse blood were measured by ELISA at the indicated time points. [Figure 21B] Figures 21A and 21B are a set of graphs showing the silencing of PCSK9 in transgenic mice expressing human PCSK9, measured over 42 days (Figure 21A) and 84 days (Figure 21B). As described in Example 11, human PCSK9 levels in the mouse blood were measured by ELISA at the indicated time points. [Figure 22] Figure 22 shows the experimental timeline for conducting a dose-response experiment to test PCSK9 silencing in transgenic mice expressing human PCSK9, as described in Example 11. [Figure 23A] Figures 23A and 23B are a set of graphs showing the results of dose-response experiments testing PCSK9 silencing in transgenic mice expressing human PCSK9. The constructs tested included the CRISPR-Off (Figure 23A) and ZF-Off (Figure 23B) constructs. [Figure 23B] Figures 23A and 23B are a set of graphs showing the results of dose-response experiments testing PCSK9 silencing in transgenic mice expressing human PCSK9. The constructs tested included the CRISPR-Off (Figure 23A) and ZF-Off (Figure 23B) constructs. [Figure 24] Figure 24 shows the results of the partial hepatectomy (PHx) experiment, demonstrating the persistence of PCSK9 epigenetic silencing in mice, as described in Example 17. [Figure 25] Figure 25 shows that multiple effectors exhibited good specificity in Huh-7 cells. Huh-7 cells were treated with different repressor fusion proteins (fusion proteins 11 and 13, respectively) and PCSK9 targeting guide RNA 041. Changes in gene expression were evaluated by RNA sequencing of treated and untreated cells, and the values ​​were visualized in a volcano plot. No significant off-target effects were detected. [Figure 26A]Figures 26A-C show the top five guides that exhibited potent PCSK9 silencing in PXB cells. Novel human hepatocytes were isolated from PXB mouse model mice. Long-term stability and functionality of the hepatocytes, as well as potent PCSK9 secretion, were confirmed. As indicated in the figure caption, cells were treated with the epigenetic repressor PLA2628 (fusion protein 12 shown in Example 12) and controls, and PCSK9 secretion was measured and plotted against negative controls (PLA2628 and non-PCSK9 targeted gRNAs). Specificity assays were performed by RNA-seq on day 14 after all gRNAs were delivered. Volcano plots for the two exemplary RNAs evaluated are shown on the right (volcano plots for others are not shown). No significant off-target effects were observed for any of the gRNAs tested. [Figure 26B] Figures 26A-C show the top five guides that exhibited potent PCSK9 silencing in PXB cells. Novel human hepatocytes were isolated from PXB mouse model mice. Long-term stability and functionality of the hepatocytes, as well as potent PCSK9 secretion, were confirmed. As indicated in the figure caption, cells were treated with the epigenetic repressor PLA2628 (fusion protein 12 shown in Example 12) and controls, and PCSK9 secretion was measured and plotted against negative controls (PLA2628 and non-PCSK9 targeted gRNAs). Specificity assays were performed by RNA-seq on day 14 after all gRNAs were delivered. Volcano plots for the two exemplary RNAs evaluated are shown on the right (volcano plots for others are not shown). No significant off-target effects were observed for any of the gRNAs tested. [Figure 26C]Figures 26A-C show the top five guides that exhibited potent PCSK9 silencing in PXB cells. Novel human hepatocytes were isolated from PXB mouse model mice. Long-term stability and functionality of the hepatocytes, as well as potent PCSK9 secretion, were confirmed. As indicated in the figure caption, cells were treated with the epigenetic repressor PLA2628 (fusion protein 12 shown in Example 12) and controls, and PCSK9 secretion was measured and plotted against negative controls (PLA2628 and non-PCSK9 targeted gRNAs). Specificity assays were performed by RNA-seq on day 14 after all gRNAs were delivered. Volcano plots for the two exemplary RNAs evaluated are shown on the right (volcano plots for others are not shown). No significant off-target effects were observed for any of the gRNAs tested. [Figure 27] Figure 27 shows the observation of secreted PCSK9 levels measured on day 14 in PXB cells. [Modes for carrying out the invention]

[0041] This disclosure provides an epigenetic editor for regulating the expression of the PCSK9 gene. By altering the expression of PCSK9, the systems, compositions, and methods described herein can be used to treat conditions such as hypercholesterolemia (e.g., heterozygous familial hypercholesterolemia (HeFH), homozygous familial hypercholesterolemia (HoFH), familial hypercholesterolemia (HF), or established atherosclerotic cardiovascular disease (ASCVD)) or renal insufficiency (RI). Unless otherwise indicated, “PCSK9” herein refers to human PCSK9. The human PCSK9 gene sequence can be found at Ensembl accession number ENSG00000169174. This epigenetic editor offers several advantages over other genomic manipulation methods, including reversibility, reduced risk of translocation, and persistent, inheritable silencing.

[0042] In some embodiments, the region of the human PCSK9 gene targeted for epigenetic regulation is approximately 2 kb in length and ± approximately 1 kb from the PCSK9 TSS. In a particular embodiment, this region has the nucleotide sequence of SEQ ID NO: 1488. In some embodiments, the targeted PCSK9 region is approximately 1069 bp in length and ± approximately 500 bp from the PCSK9 TSS. In a particular embodiment, the targeted region has the nucleotide sequence of SEQ ID NO: 1489. The PCSK9 TSS is located at #chr1:55039548 in genomic GRCh38.

[0043] In some embodiments, the epigenetic editors described herein may comprise one or more fusion proteins, each comprising a DNA-binding domain ligated to one or more effector domains for epigenetic modification. In certain embodiments, if the DNA-binding domain is a DNA-binding domain guided to a polynucleotide, the epigenetic editor may further comprise one or more guide polynucleotides. The DNA-binding domains, effector domains, and guide polynucleotides of the epigenetic editors described herein may be selected in any functional combination from, for example, those described below.

[0044] The epigenetic editors described herein may be transiently expressed in host cells or integrated into the genome of host cells, and such cells and their offspring are also intended by this disclosure. Both transiently expressed and integrated epigenetic editors, or their components, can result in stable epigenetic modifications. For example, after introducing an epigenetic editor described herein into host cells, target genes in the host cells can be stably or permanently suppressed or silenced. In some embodiments, the expression of the target gene is reduced or silenced for at least one week, at least two weeks, at least three weeks, at least four weeks, at least five weeks, at least six weeks, at least seven weeks, at least two months, at least three months, at least four months, at least five months, at least six months, at least one year, at least two years, or for the entire lifespan of the cell or the object holding the cell, compared to the expression level in the absence of the epigenetic editor. The epigenetic modifications may be inherited by the offspring of the host cells into which the epigenetic editor was introduced.

[0045] This epigenetic editor can be introduced into patients who need it (e.g., human patients), such as patient hepatocytes, cholangiocarcinoma (cholangiocarcinoma), stellate cells, Kupffer cells, and hepatic sinusoidal endothelial cells.

[0046] I. DNA-binding domain The epigenetic editors described herein may include one or more DNA-binding domains that direct the effector domain(s) of the epigenetic editor to a target sequence within or near the PCSK9 locus. The DNA-binding domains described herein may include, for example, polynucleotide-guided DNA-binding domains, zinc finger protein (ZFP) domains, transcription activator-like effector (TALE) domains, and meganuclease DNA-binding domains. An example of a DNA-binding domain can be found in U.S. Patent No. 11,162,114, which is incorporated herein by reference in its entirety.

[0047] In some embodiments, the DNA-binding domain described herein is encoded by its native coding sequence. In other embodiments, the DNA-binding domain is encoded by a nucleotide sequence that has been codon-optimized for optimal expression in human cells.

[0048] A. DNA-binding domain guided by polynucleotides In some embodiments, the DNA-binding domain as described herein may be a protein domain directed to a target site at the PCSK9 locus by a guide nucleic acid sequence (e.g., a guide RNA sequence). In certain embodiments, the protein domain may be derived from a CRISPR-related nuclease, e.g., a class I or II CRISPR-related nuclease. In some embodiments, the protein domain may be derived from a Cas nuclease, e.g., type II, type IIA, type IIB, type IIC, type V, or type VI Cas nuclease. In a particular embodiment, the protein domains are Cas1, Cas1B, Cas2, Cas3, Cas4, Cas5, Cas6, Cas7, Cas8, Cas9, Cas10, Cas14a, Cas14b, Cas14c, CasX, CasY, CasPhi, C2c4, C2c8, C2c9, C2c10, Csy1, Csy2, Csy3, Cse1, Cse2, Csc1, Csc2, Csa5 The protein may be derived from a class II Cas nuclease selected from Csn2, Csm2, Csm3, Csm4, Csm5, Csm6, Cmr1, Cmr3, Cmr4, Cmr5, Cmr6, Csb1, Csb2, Csb3, Csx17, Csx14, Csx10, Csx16, CsaX, Csx3, Csx1, Csx1S, Csf1, Csf2, CsO, Csf4, and their homologs and variants. "Derived from" is used to mean that the protein domain contains the full-length polypeptide sequence of the parent protein or contains a variant thereof (e.g., having deletions, insertions, and / or substitutions of amino acid residues). The variant retains the desired function of the parent protein (e.g., the ability to form a guide nucleic acid sequence and a complex with target DNA).

[0049] In some embodiments, the CRISPR-associated protein domain may be a Cas9 domain as described herein. Cas9 may refer to a polypeptide having at least about 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity and / or similarity to a wild-type Cas9 polypeptide as described herein. In some embodiments, the wild-type polypeptide is Cas9 derived from Streptococcus pyogenes (NCBI reference number NC_002737.2 (SEQ ID NO: 1)) and / or UniProt reference number Q99ZW2 (SEQ ID NO: 2). In some embodiments, the wild-type polypeptide is Cas9 derived from Staphylococcus aureus (SEQ ID NO: 3). In some embodiments, the CRISPR-associated protein domain is a polypeptide having at least about 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity and / or similarity to the Cpf1 domain or protein, or to the wild-type Cpf1 polypeptide described herein (e.g., Cpf1 derived from Franscisella novicida (UniProt reference number U2UMQ6 or SEQ ID NO: 4)). In certain embodiments, the CRISPR-associated protein domain may be a modified form, fusion, or chimera of the wild-type protein, including changes to one or more amino acid residues, e.g., deletions, insertions, or substitutions, or any combination thereof.

[0050] The structures of Cas9 sequences and variant Cas9 orthologues have been described for various organisms. Exemplary organisms from which the Cas9 domain may originate in this specification include Streptococcus pyogenes, Streptococcus thermophilus, Streptococcus sp., Staphylococcus aureus, Listeria innocua, Lactobacillus gasseri, Francisella nobicida, Wolinella succinogenes, Sutterella wadsworthensis, Gamma proteobacterium, Neisseria meningitidis, Campylobacter jejuni, and Pasteurella murtofida. multocida), Fibrobacter succinogene, Rhodospirillum rubrum, Nocardiopsis dassonvillei, Streptomyces pristinaespiralis, Streptomyces viridochromogenes, Streptomyces viridochromogenes, Streptomyces roseum, Alicyclobacillus acidocaldarius, Bacillus pseudomycoides, Bacillus serenityledacens Exiguobacterium selenitireducens, Exiguobacterium sibiricumLactobacillus sibiricum, Lactobacillus delbrueckii, Lactobacillus salivarius, Lactobacillus buchneri, Treponema denticola, Microscilla marina, Burkholderiales bacterium, Polaromonas naphthalenivorans, Polaromonas sp., Crocosphaera watsonii, Cyanothece sp., Microcystis aeruginosa, Synechococcus sp.), Acetohalobium arabaticum, Ammonifex degensii, Caldicelulosiruptor becscii, Candidatus Desulforudis, Clostridium botulinum, Clostridium difficile, Finegoldia magna, Natranaerobius thermophilus, Pelotomaculum thermopropionium, Acidithiobacillus cardus Caldus), Acidithiobacillus ferrooxidans, Allochromatium vinosum, Marinobacter sp., Nitrosococcus halophilushalophilus), Nitrosococcus watsoni, Pseudoalteromonas haloplanktis, Ktedonobacter racemifer, Methanohalobium evestigatum, Anabaena variabilis, Nodularia spumigena, Nostoc sp., Arthrospira maxima, Arthrospira platensis, Arthrospira sp., Lyngbya sp., Microcoleus xonoplastes Examples of Cas9 sequences include, but are not limited to, those derived from organisms and loci disclosed by Chylinski et al., RNA Biol. (2013) 10(5):726-37.

[0051] In some embodiments, the Cas9 domain is derived from Streptococcus pyogenes (SpCas9). In some embodiments, the Cas9 domain is derived from Staphylococcus aureus (SaCas9).

[0052] Other Cas domains are also intended for use in the epigenetic editor herein. These include, for example, those derived from CasX (Cas12E) (e.g., SEQ ID NO: 5), CasY (Cas12d) (e.g., SEQ ID NO: 6), Casφ (CasPhi) (e.g., SEQ ID NO: 7), Cas12f1 (Cas14a) (e.g., SEQ ID NO: 8), Cas12f2 (Cas14b) (e.g., SEQ ID NO: 9), Cas12f3 (Cas14c) (e.g., SEQ ID NO: 10), and C2c8 (e.g., SEQ ID NO: 11).

[0053] With respect to epigenetic editing, the protein domain derived from the nuclease (e.g., the Cas9 or Cpf1 domain) may have reduced nuclease activity through mutation, or may not have nuclease activity at all, so that the protein domain does not cleave DNA or has reduced DNA cleavage activity, while simultaneously retaining the ability to form a complex with a guide nucleic acid sequence (e.g., guide RNA) and target DNA. For example, the nuclease activity may be reduced by at least 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 99% compared to the wild-type domain. In some embodiments, the CRISPR-related protein domains described herein are catalytically inactive ("dead"). Examples of such domains include, for example, dCas9 ("dead" Cas9), dCpf1, ddCpf1, dCasPhi, ddCas12a, dLbCpf1, and dFnCpf1. The dCas9 protein domain may contain one, two, or more mutations that suppress its nuclease activity compared to, for example, wild-type Cas9. The DNA cleavage domain of Cas9 is known to contain two subdomains: the HNH nuclease subdomain and the RuvC1 subdomain. The HNH subdomain cleaves the strand complementary to the gRNA, while the RuvC1 subdomain cleaves the non-complementary strand. Mutations within these subdomains can silence the nuclease activity of Cas9. For example, the mutations D10A (in the case of RuvC1) and H840A (in the case of HNH) completely inactivate the nuclease activity of SpCas9. Similarly, SaCas9 can also be inactivated by mutations D10A and N580A. In some embodiments, dCas9 contains at least one mutation in the HNH subdomain and / or RuvC1 subdomain that reduces or inhibits nuclease activity. In some embodiments, dCas9 contains either the RuvC1 subdomain alone or the HNH subdomain alone.It should be understood that any mutations that inactivate the RuvC1 and / or HNH domains, such as insertions, deletions, or single or multiple amino acid substitutions in the RuvC1 and / or HNH domains, may be included in dCas9 as used herein.

[0054] In some embodiments, the dCas9 protein herein contains mutations at positions corresponding to the D10 (e.g., D10A), H840 (e.g., H840A), or both, as numbered in the sequence provided by UniProt accession number Q99ZW2 (SEQ ID NO: 2). In certain embodiments, dCas9 comprises the amino acid sequence of dSpCas9 (D10A and H840A) (SEQ ID NO: 12).

[0055] In some embodiments, the dCas9 protein described herein contains mutations at positions corresponding to the D10 position (e.g., D10A), the N580 position (e.g., N580A), or both, of the wild-type SaCas9 sequence (e.g., SEQ ID NO: 3). In certain embodiments, dCas9 contains the amino acid sequence of dSaCas9 (D10A and N580A) (SEQ ID NO: 13).

[0056] Additional suitable mutations for inactivating Cas9 will be apparent to those skilled in the art based on the present disclosure and knowledge of the art, and are within the scope of the present disclosure. Such mutations include, but are not limited to, D839A, N863A, and / or K603R in SpCas9. The present disclosure intends to provide any mutation (for example, a mutation corresponding to any of the Cas9 mutations described herein) that reduces or inhibits the nuclease activity of any Cas9 described herein.

[0057] The dCpf1 protein domain may contain one, two, or more mutations that reduce or inhibit its nuclease activity compared to wild-type Cpf1. The Cpf1 protein has a RuvC-like endonuclease domain similar to the RuvC domain of Cas9, but lacks an HNH endonuclease domain, and the N-terminus of Cpf1 does not have the alpha-helix recognition lobe of Cas9. In some embodiments, dCpf1 contains one or more mutations corresponding to the numbered positions D917A, E1006A, or D1255A in the sequence of the Francisella nobicida Cpf1 protein (FnCpf1, SEQ ID NO: 4). In certain embodiments, the dCpf1 protein includes mutations corresponding to positions D917A, E1006A, D1255A, D917A / E1006A, D917A / D1255A, E1006A / D1255A, or D917A / E1006A / D1255A, or corresponding mutations (may be multiple) in any of the Cpf1 amino acid sequences described herein. In some embodiments, dCpf1 includes the D917A mutation. In certain embodiments, dCpf1 includes the amino acid sequence of dFnCpf1 (SEQ ID NO: 14).

[0058] Further nuclease-inactive CRISPR-related protein domains intended herein include, for example, those derived from dNmeCas9 (e.g., SEQ ID NO: 15), dCjCas9 (e.g., SEQ ID NO: 16), dSt1Cas9 (e.g., SEQ ID NO: 17), dSt3Cas9 (e.g., SEQ ID NO: 18), dLbCpf1 (e.g., SEQ ID NO: 19), dAsCpf1 (e.g., SEQ ID NO: 20), denAsCpf1 (e.g., SEQ ID NO: 21), dHFAsCpf1 (e.g., SEQ ID NO: 22), dRVRAsCpf1 (e.g., SEQ ID NO: 23), dRRAsCpf1 (e.g., SEQ ID NO: 24), dCasX (e.g., SEQ ID NO: 25), and dCasPhi (e.g., SEQ ID NO: 26).

[0059] In some embodiments, the Cas9 domains described herein may be high-fidelity Cas9 domains comprising, for example, one or more mutations that reduce the electrostatic interaction between the Cas9 domain and the sugar-phosphate backbone of DNA to confer increased target binding specificity. In certain embodiments, the high-fidelity Cas9 domain may be nuclease-inactive as described herein.

[0060] The CRISPR-related protein domains described herein can recognize protospacer-adjacent motif (PAM) sequences in target genes. A "PAM" sequence is typically a 2-6 bp DNA sequence immediately following the sequence targeted by the CRISPR-related protein domain. PAM sequences are required for CRISPR protein binding and cleavage, but are not part of the target sequence. CRISPR-related protein domains can recognize naturally occurring or classical PAM sequences, or they may have modified PAM specificity. CRISPR-related protein domains that bind to non-classical PAM sequences have been described in the art. For example, Cas9 domains that bind to non-classical PAM sequences are described in Kleinstiver et al., Nature (2015) 523(7561):481-5 and Kleinstiver et al., Nat Biotechnol. (2015) 33:1293-8. Examples of such Cas9 domains include those derived from "VRER" SpCas9, "EQR" SpCas9, "VQR" SpCas9, "SpG Cas9", "SpRYCas9", and "KKH" SaCas9. Nuclease-inactive forms of these Cas9 domains are also intended, such as nuclease-inactive VRER SpCas9 (e.g., SEQ ID NO: 27), nuclease-inactive EQR SpCas9 (e.g., SEQ ID NO: 28), nuclease-inactive VQR SpCas9 (e.g., SEQ ID NO: 29), nuclease-inactive SpG Cas9 (e.g., SEQ ID NO: 30), nuclease-inactive SpRY Cas9 (e.g., SEQ ID NO: 31), and nuclease-inactive KKH SaCas9 (e.g., SEQ ID NO: 32). Another example is the Cas9 of Francisella nobicida, which has been engineered to recognize 5'-YG-3' (where "Y" is a pyrimidine).

[0061] Additional suitable CRISPR-related proteins, orthologues, and variants including nuclease-inactive variants and sequences will be apparent to those skilled in the art based on this disclosure.

[0062] Guide RNAs that can be used with CRISPR-related protein domains in this specification are further described in Section II below.

[0063] B. Zinc finger protein domain In some embodiments, the DNA-binding domain of the epigenetic editor described herein comprises a zinc finger protein (ZFP) domain (or, as used herein, a “ZF domain”). A ZFP is a protein having at least one zinc finger and binds to DNA in a sequence-specific manner. A “zinc finger” (ZF) or “zinc finger motif” (ZF motif) refers to a polypeptide domain containing a beta-beta-alpha (ββα) type protein fold stabilized by a zinc ion. A ZF binds to 2 to 4 nucleotide base pairs, typically 3 or 4 base pairs (continuous or discontinuous). Each ZF typically contains about 30 amino acids. A ZFP domain may contain multiple ZFs in tandem contact with their target nucleic acid sequences. A tandem array of ZFs may be engineered to generate artificial ZFPs that bind to a desired nucleic acid target. ZFPs can be reasonably designed using a database containing triplet (or quadruplet) nucleotide sequences and individual ZF amino acid sequences, where each triplet or quadruplet nucleotide sequence is associated with one or more amino acid sequences of ZFs that bind to a specific triplet or quadruplet sequence. See, for example, U.S. Patents 6,453,242, 6,534,261, and 8,772,453.

[0064] ZFPs are widely present in eukaryotic cells and may belong to classes such as the C2H2 class, CCHC class, PHD class, or RING class. An exemplary motif characterizing one class of these proteins (the C2H2 class) is -Cys-(X) 2-4 -Cys-(X) 12 -His-(X) 3-5-His-(SEQ ID NO: 657), where X is any independently selected amino acid. In some embodiments, the ZFP domains herein may comprise a ZF array containing a sequence of C2H2-ZFs, each in contact with three or more consecutive nucleotides.

[0065] The ZFP domains of the epigenetic editors described herein may contain two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, or more ZFs. The ZFP domain may contain an array of two-finger or three-finger units, for example, three, four, five, six, seven, eight, nine, or ten, or more units, where each unit binds to a subsite in the target sequence. In some embodiments, a ZFP domain containing at least three ZFs recognizes a target DNA sequence of nine or ten nucleotides. In some embodiments, a ZFP domain containing at least four ZFs recognizes a target DNA sequence of 12 to 14 nucleotides. In some embodiments, a ZFP domain containing at least six ZFs recognizes a target DNA sequence of 18 to 21 nucleotides.

[0066] In some embodiments, the ZFs in the ZFP domain described herein are linked via peptide linkers. The peptide linkers may be, for example, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or more amino acids long. In some embodiments, the linker contains 5 or more amino acids. In some embodiments, the linker contains 7 to 17 amino acids. The linker may be mobile or rigid.

[0067] In some embodiments, the zinc finger array is arranged as follows: The sequence may have TIFF2026510339000002.tif25162 or a sequence that is at least 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% identical thereto, where "XXXXXXX" represents the amino acids of the ZF-recognizing helix that confers DNA-binding specificity to the zinc finger, and each X can be independently selected. In the above sequence, the italicized "XX" may be TR, LR, or LK, and "[linker]" represents the linker sequence. In some embodiments, the linker sequence is TGSQKP (sequence ID 651), which can be used when the subsites targeted by the ZF are adjacent. In some embodiments, the linker sequence is TGGGGSQKP (sequence ID 652), which can be used when there are bases between the subsites targeted by the zinc finger. The two linkers shown may be the same or different.

[0068] In this specification, a ZFP domain may comprise an array of two or more adjacent ZFs that are directly adjacent to one another (e.g., separated by a short (classical) linker sequence) or separated by a long mobile or structured polypeptide sequence. In some embodiments, directly adjacent fingers bind to a contiguous nucleic acid sequence, i.e., adjacent trinucleotides / triplets. In some embodiments, adjacent fingers cross-link between their respective target triplets, which may help enhance or augment the recognition of the target sequence, resulting in the binding of overlapping sequences. In some embodiments, distal ZFs within a ZFP domain may recognize (or bind to) discontinuous nucleotide sequences.

[0069] The amino acid sequences of the ZF DNA recognition helices of exemplary ZFP domains and their PCSK9 target sequences are shown in Table 1 below, where the numbers in parentheses are the sequence numbers. Table 1. Exemplary ZFP domain ZF sequences targeting PCSK9

[0070] [Table 1] TIFF2026510339000004.tif255167TIFF2026510339000005.tif145168

[0071] In some embodiments, the ZFP domain of this epigenetic editor binds to a target sequence selected from any one of sequence numbers 700 to 747. In further embodiments, the ZFP domain sequentially comprises any one of the F1 to F6 amino acid sequences from ZF001 to ZF048 shown in Table 1. The F1 to F6 amino acid sequences may be placed within the ZF framework sequence of sequence number 650, or within any other ZF framework known in the art.

[0072] C.TALE In some embodiments, the DNA-binding domain of the epigenetic editor described herein includes a transcription activator-like effector (TALE) domain. The DNA-binding domain of the TALE comprises a highly conserved sequence of approximately 33–34 amino acids and has two repeat variable di-residues (RVDs) at positions 12 and 13 that are central to the recognition of a specific nucleotide. The TALE can be manipulated to bind to virtually any desired DNA sequence. Methods for programming the TALE are known in the art. For example, such methods are described in Carroll et al., Genet Soc Amer. (2011) 188(4):773-82, Miller et al., Nat Biotechnol. (2007) 25(7):778-85, Christian et al., Genetics (2008) 186(2):757-61, Li et al., Nucl Acids Res. (2010) 39(1):359-72, and Moscou et al., Science (2009) 326(5959):1501.

[0073] D. Other DNA-binding domains Other DNA-binding domains are intended for the epigenetic editors described herein. In some embodiments, the DNA-binding domains include those derived from Argonaut protein domains, e.g., Natronobacterium gregoryi (NgAgo). NgAgo is an ssDNA-guided endonuclease that is guided by 5' phosphorylated ssDNA (gDNA) to its target site, where it results in a double-strand break. In contrast to Cas9, the NgAgo-gDNA system does not require a protospacer adjacent motif (PAM). Therefore, by using nuclease-inactive NgAgo (dNgAgo), the range of bases that can be targeted can be greatly expanded. The characterization and use of NgAgo are described, for example, in Gao et al., Nat Biotechnol. (2016) 34(7):768-73, Swarts et al., Nature (2014) 507(7491):258-61, and Swarts et al., Nucl Acids Res. (2015) 43(10):5120-9.

[0074] In some embodiments, the DNA-binding domain includes an inactivated nuclease, such as an inactivated meganuclease. Additional non-limiting examples of DNA-binding domains include a tetracycline-regulated repressor (tetR) DNA-binding domain, a leucine zipper, a helix-loop-helix (HLH) domain, a helix-turn-helix domain, a β-sheet motif, a steroid receptor motif, a bZIP domain, a homeodomain, and an AT hook.

[0075] II. Guide Polynucleotides An epigenetic editor described herein, comprising a DNA-binding domain guided to a polynucleotide, may also include a guide polynucleotide capable of forming a complex with the DNA-binding domain. The guide polynucleotide may include RNA, DNA, or a mixture of both. For example, if the DNA-binding domain guided to the polynucleotide is a CRISPR-associated protein domain, the guide polynucleotide may be a guide RNA (gRNA). "Guide RNA" or "gRNA" refers to a nucleic acid capable of hybridizing to a target sequence and inducing the binding of a CRISPR-Cas complex to the target sequence. Methods of using a guide polynucleotide sequence in conjunction with a programmable DNA-binding protein (e.g., a CRISPR-associated protein domain) for site-specific DNA targeting (e.g., to modify a genome) are known in the art.

[0076] A guide polynucleotide sequence (e.g., a gRNA sequence) may comprise two parts: 1) a nucleotide sequence containing a target nucleic acid sequence ("target sequence"), e.g., a "targeting sequence" complementary to a nucleic acid sequence located in a genomic target site; and 2) a nucleotide sequence that binds to a DNA-binding domain (e.g., a CRISPR-Cas protein domain) guided by the polynucleotide. The nucleotide sequence in 1) may contain a targeting sequence that is 100% complementary to a genomic nucleic acid sequence, e.g., a nucleic acid sequence located in a genomic target site, and is therefore capable of hybridizing to the target nucleic acid sequence. The nucleotide sequence in 1) may be referred to as, for example, crispr RNA or crRNA. The nucleotide sequence in 2) may be referred to as a scaffold sequence of the guide nucleic acid, e.g., tracrRNA, or an activation region of the guide nucleic acid, and may comprise a stem-loop structure. The parts 1) and 2) described above may fused to form a single guide (e.g., a single guide RNA or sgRNA) or may be located on two separate nucleic acid molecules. In some embodiments, the guide polynucleotide includes portions 1) and 2) linked by a linker. In some embodiments, the guide polynucleotide includes portions 1) and 2) linked by a non-nucleic acid linker, such as a peptide linker or a chemical linker.

[0077] The portion 2 (scaffold sequence) of the guide polynucleotide described herein may be, for example, those described in Jinek et al., Science (2012) 337:816-21, U.S. Patent Application Publication No. 2016 / 0208288, or U.S. Patent Application Publication No. 2016 / 0200779. Variants of portion 2) are also contemplated in this disclosure. For example, the tetraloop and stemloop of the gRNA scaffold (tracrRNA) sequence may be modified to include an RNA aptamer that can be bound by a specific protein domain. In some embodiments, such a modified gRNA can be used to facilitate the recruitment of a repressive or activating domain fused to an RNA aptamer that interacts with a protein.

[0078] The gRNAs provided herein typically include a targeting domain and a binding domain. The targeting domain (also referred to as the “targeting sequence”) may include a target site, for example, a nucleic acid sequence that binds to a genomic nucleic acid molecule within a cell. The target site may be a double-stranded DNA sequence containing a PAM sequence and a target sequence located on the same strand as the PAM sequence, directly adjacent to it. The targeting domain of a gRNA may include an RNA sequence corresponding to the target sequence; that is, it is similar to the sequence of the target domain and may have one or more mismatches, but typically contains an RNA sequence instead of a DNA sequence. The targeting domain of a gRNA can therefore base-pair (with complete or partial complementarity) with the sequence of a double-stranded target site complementary to the target sequence, and thus with the strand complementary to the strand containing the PAM sequence. It will be understood that the targeting domain of a gRNA typically does not contain a sequence similar to the PAM sequence. It will be further understood that the position of the PAM may be 5' or 3' of the target sequence, depending on the nuclease used. For example, the PAM is typically located at 3' of the target sequence for Cas9 nucleases and at 5' for Cas12a nucleases. For an example of the PAM location and the mechanism by which gRNA binds to the target site, see, for example, Figure 1 in Vanegas et al., Fungal Biol Biotechnol. (2019) 6:6, which is incorporated herein by reference. For additional examples and explanations of the mechanism of gRNA targeting of target sites by RNA-guided nucleases, see Fu et al., Nat Biotechnol (2014) 32(3):279-84 and Sternberg et al., Nature (2014) 507(7490):62-7, which are incorporated herein by reference, respectively.

[0079] In some embodiments, the targeting domain sequence contains 17 to 30 nucleotides and perfectly corresponds to the target sequence (i.e., it has no mismatched nucleotides). In some embodiments, however, the targeting domain sequence may contain one or more, typically four or fewer, mismatches, e.g., one, two, three, or four mismatches. The targeting domain is part of a gRNA, which is an RNA molecule and therefore typically contains ribonucleotides, while a DNA targeting domain would contain deoxyribonucleotides.

[0080] An illustrative diagram of a gRNA containing a Cas9 target site with 22 nucleotide target domains and an NGG PAM sequence, as well as a targeting domain that perfectly corresponds to the target sequence (and therefore base-pairs with the DNA strand complementary to the strand containing the target sequence and PAM in perfect complementarity), is provided below. TIFF2026510339000006.tif37168

[0081] An illustrative diagram of a gRNA containing a Casl2a target site with 22 nucleotide target domains and a TTN PAM sequence, as well as a targeting domain that perfectly corresponds to the target sequence (and therefore base-pairs with the DNA strand complementary to the strand containing the target sequence and PAM in perfect complementarity), is provided below. TIFF2026510339000007.tif37168

[0082] While we do not wish to be constrained by theory, in at least some embodiments, the lengths of the targeting domain and the target sequence, and their complementarity, are thought to contribute to the specificity of the interaction between the gRNA / Cas9 molecular complex and the target nucleic acid. In some embodiments, the targeting domain of the gRNA provided herein is 5 to 50 nucleotides long. In some embodiments, the targeting domain is 15 to 25 nucleotides long. In some embodiments, the targeting domain is 18 to 22 nucleotides long. In some embodiments, the targeting domain is 19 to 21 nucleotides long. In some embodiments, the targeting domain is 15 nucleotides long. In some embodiments, the targeting domain is 16 nucleotides long. In some embodiments, the targeting domain is 17 nucleotides long. In some embodiments, the targeting domain is 18 nucleotides long. In some embodiments, the targeting domain is 19 nucleotides long. In some embodiments, the targeting domain is 20 nucleotides long. In some embodiments, the targeting domain is 21 nucleotides long. In some embodiments, the targeting domain is 22 nucleotides long. In some embodiments, the targeting domain is 23 nucleotides long. In some embodiments, the targeting domain is 24 nucleotides long. In some embodiments, the targeting domain is 25 nucleotides long. In certain embodiments, the targeting domain corresponds perfectly to the target sequence or a portion thereof provided herein without any mismatches. In some embodiments, the targeting domain of the gRNA provided herein contains one mismatch with respect to the target sequence provided herein. In some embodiments, the targeting domain contains two mismatches with respect to the target sequence. In some embodiments, the target domain contains three mismatches with respect to the target sequence.

[0083] Methods for designing, selecting, and validating gRNAs are described herein and are publicly known in the art. Software tools can be used to optimize gRNAs corresponding to target DNA sequences, for example, to minimize total off-target activity across the genome. For example, DNA sequence search algorithms can be used to identify target sequences within the crRNA of a gRNA for use with Cas9. Exemplary gRNA design tools are described in Bae et al., Bioinformatics (2014) 30:1473-5.

[0084] The guide polynucleotides (e.g., gRNAs) described herein may be of various lengths. In some embodiments, the length of the spacer or targeting sequence depends on the CRISPR-related protein component of the epigenetic editor system used. For example, Cas proteins derived from different bacterial species have a variety of optimal targeting sequence lengths. Therefore, the spacer sequence may contain lengths of, for example, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, or more than 50 nucleotides. In some embodiments, the spacer has a length of 10-24, 11-20, 11-16, 18-24, 19-21, or 20 nucleotides. In some embodiments, the guide polynucleotide (e.g., gRNA) has 15-100 nucleos (e.g., 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50) nucleos The guide polynucleotide is of a length of nucleotide and includes a spacer sequence of at least 10 consecutive nucleotides (e.g., 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50) that are complementary to the target sequence. In some embodiments, the guide polynucleotide described herein may have, for example, 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 40, 50, or more nucleotides truncated.

[0085] In certain embodiments, the 3' end of the PCSK9 target sequence is directly adjacent to a PAM sequence (e.g., a classical PAM sequence, e.g., NGG for SpCas9). The degree of complementarity between the targeting sequence of the guide polynucleotide (e.g., a spacer sequence of the gRNA) and the target sequence may be at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%. In certain embodiments, the targeting sequence and the target sequence may be 100% complementary. In other embodiments, the targeting sequence and the target sequence may contain, for example, one, two, three, four, five, six, seven, eight, nine, or ten mismatches.

[0086] Guide polynucleotides (e.g., gRNAs) may be modified, for example, by chemical modification and synthetic modification. Modified gRNAs may include, for example, alteration or substitution of one or both of the unbound phosphate oxygens and / or one or more of the bound phosphate oxygens in the phosphodiester backbone linkage, alteration of the ribose sugar (e.g., the 2' hydroxyl group of the ribose sugar), alteration of the phosphate moiety, alteration or substitution of naturally occurring nucleic acid bases, alteration or substitution of the ribose-phosphate backbone, alteration of the 3' and / or 5' ends of the oligonucleotide, substitution of terminal phosphate groups, or conjugation of a part, cap, or linker, or any combination thereof.

[0087] In some embodiments, one or more ribose groups of the gRNA may be modified. Examples of chemical modifications to ribose groups include, but are not limited to, 2'-O-methyl (2'-OMe), 2'-fluoro (2'-F), 2'-deoxy, 2'-O-(2-methoxyethyl) (2'-MOE), 2'-NH2, 2'-O-allyl, 2'-O-ethylamine, 2'-O-cyanoethyl, 2'-O-acetal ester, or bicyclic nucleotides such as locked nucleic acid (LNA), 2'-(5-restricted ethyl (S-cEt)), restricted MOE, or 2'-0,4'-C-aminomethylene-bridged nucleic acid (2',4'-BNANC). 2'-O-methyl and / or 2'-fluoro modifications may increase the binding affinity and / or nuclease stability of the gRNA oligonucleotide.

[0088] In some embodiments, one or more phosphate groups of the gRNA may be chemically modified. Examples of chemical modifications to phosphate groups include, but are not limited to, phosphorothioate (PS), phosphonoacetate (PACE), thiophosphonoacetate (thioPACE), amide, triazole, phosphonate, and phosphotryester modifications. In some embodiments, the guide polynucleotides described herein may contain one, two, three, or more PS ligatures at or near the 5' and / or 3' ends, and the PS ligatures may be continuous or discontinuous.

[0089] In some embodiments, the gRNA as used herein comprises a mixture of ribonucleotides and deoxyribonucleotides, and / or one or more PS linkages.

[0090] In some embodiments, one or more nucleic acid bases of the gRNA may be chemically modified. Examples of chemically modified nucleic acid bases include, but are not limited to, 2-thiouridine, 4-thiouridine, N6-methyladenosine, pseudouridine, 2,6-diaminopurine, inosine, thymidine, 5-methylcytosine, 5-substituted pyrimidine, isoguanine, isocytosine, and nucleic acid bases having halogenated aromatic groups. Chemical modification may be performed on the spacer region, the tracr RNA region, the stem-loop, or any combination thereof.

[0091] Table 2 below lists exemplary gRNA target sequences for epigenetic modification of human PCSK9, as well as the coordinates of the start and end locations of the targeted sites on human chromosome 1 (SEQ: Sequence ID). This table also shows the distance from the start coordinate to the TSS coordinate of the PCSK9 gene. Table 2. Exemplary target sequences of gRNAs that target PCSK9

[0092] [Table 2] TIFF2026510339000009.tif255167TIFF2026510339000010.tif255167TIFF2026510339000011.tif255167TIFF2026510339000012.tif255167

[0093] In some embodiments, the gRNAs described herein do not contain the sequence CCCGCACCUUGGCGCAGCGG (SEQ ID NO: 1490).

[0094] Any tracr sequence known in the art is intended for the gRNA described herein. In some embodiments, the gRNA described herein has a tracr sequence (SEQ: sequence number) that is at least 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% identical to the tracr sequences shown below in Table 3. Table 3. Exemplary TRACR sequences

[0095] [Table 3]

[0096] In some embodiments, the gRNAs described herein are delivered directly to cells (e.g., via an RNP complex together with a CRISPR-related protein domain). In some embodiments, the gRNAs are delivered to cells via an expression vector (e.g., a plasmid vector or a viral vector) into which the cells are introduced, and the cells then express the gRNAs from the expression vector. Methods for introducing gRNAs and expression vectors into cells are well known in the art.

[0097] III. Effector Domain The epigenetic editors described herein comprise one or more effector protein domains (also referred to herein as “epigenetic effector domains” or “effector domains”) that result in epigenetic modification of a target gene. Epigenetic editors having one or more effector domains can regulate the expression of a target gene without altering the nucleic acid sequence of the target gene. In some embodiments, the effector domains described herein may provide repression or silencing of the expression of a target gene, such as PCSK9, for example, by repressing transcription or by modifying or remodeling chromatin. Such effector domains are also referred herein as “repressor domains,” “repressor domains,” or “epigenetic repressor domains.” Non-limiting examples of chemical modifications that can be mediated by effector domains include methylation, demethylation, acetylation, deacetylation, phosphorylation, SUMOylation, and / or ubiquitination of DNA or histone residues.

[0098] In some embodiments, the effector domain of the epigenetic editor described herein may perform histone tail modification, for example, by adding or removing an activation mark to the histone tail.

[0099] In some embodiments, the effector domain of the epigenetic editor described herein may include or recruit transcription-related proteins, such as transcriptional repressors. These transcription-related proteins may be endogenous or exogenous.

[0100] In some embodiments, the effector domain of the epigenetic editor described herein may include, for example, a protein that directly or indirectly blocks access of a transcription factor to a target gene having a target sequence.

[0101] Effector domains can be full-length proteins or fragments thereof that retain epigenetic effector function ("functional domains"). Functional domains capable of regulating (e.g., repressing) gene expression may originate from larger proteins. For example, functional domains capable of reducing target gene expression may be identified based on the sequence of a repressor protein. The amino acid sequences of proteins that regulate gene expression can be obtained from available genome browsers, such as the UCSD genome browser or the Ensembl genome browser. Functional domains within full-length protein sequences can be identified using protein annotation databases, such as UniProt or Pfam. As a starting point, the largest sequence encompassing all regions identified by different databases can be tested for gene expression regulatory activity. Various truncated forms can then be tested to identify the smallest functional units.

[0102] Variants of the effector domains described herein are also contemplated by this disclosure. A variant refers to a polypeptide having at least about 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity and / or similarity to the wild-type effector domain described herein. In certain embodiments, a variant retains at least about 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% of the epigenetic effector function of the wild-type effector domain.

[0103] In some embodiments, the effector domains described herein may include a fusion of two or more effector domains (e.g., KOX1 KRAB and ZIM3). An effector domain may include, for example, two, three, four, five, six, seven, eight, nine, or ten effector domains, e.g., a fusion of the effector domains described herein. In a particular embodiment, an effector domain includes a truncated form of one effector domain and a fusion of a second effector domain. In a particular embodiment, an effector domain includes a fusion of truncated forms of two effector domains (e.g., a fusion of the N-terminal and C-terminal portions of two effector domains).

[0104] In some embodiments, the epigenetic editors described herein may comprise one effector domain, two effector domains, three effector domains, four effector domains, five effector domains, six effector domains, seven effector domains, eight effector domains, nine effector domains, ten effector domains, or more. In certain embodiments, the epigenetic editor comprises one or more fusion proteins (e.g., one, two, or three fusion proteins) each having one or more effector domains (e.g., one, two, or three effector domains) linked to a DNA-binding domain. In some embodiments, the effector domains may induce combinations of epigenetic modifications, such as transcriptional repression and DNA methylation, DNA methylation and histone deacetylation, DNA methylation and histone demethylation, DNA methylation and histone methylation, DNA methylation and histone phosphorylation, DNA methylation and histone ubiquitination, or DNA methylation and histone SUMOylation.

[0105] In certain embodiments, the effector domains described herein (e.g., DNMT3A and / or DNMT3L) are encoded by nucleotide sequences found in the natural genome (e.g., human or mouse) for that effector domain. In other embodiments, the effector domains described herein are encoded by nucleotide sequences that are codon-optimized for optimal expression in human cells.

[0106] The effector domains described herein may include, for example, transcriptional repressors, DNA methyltransferases, and / or histone modifiers, as will be further detailed below.

[0107] A. Transfer Repressor In some embodiments, the epigenetic effector domains described herein mediate the repression of the expression (e.g., transcription) of a target gene. Effector domains may include, for example, the Kruppel-associated box (KRAB) repressor domain, the repressor element silencing transcription factor (REST) ​​repressor domain, the KRAB-associated protein 1 (KAP1) domain, the MAD domain, the FKHR (forkhead gene in rhabdosarcoma) repressor domain, the EGR-1 (early growth response gene product-1) repressor domain, the ets2 repressor factor repressor domain (ERD), the MAD smSIN3 interaction domain (SID), the WRPW motif of the hairy-associated basic helix-loop-helix (bHLH) repressor protein, the HP1 alpha-chromoshadow repressor domain, the HP1 beta-repressor domain, or any combination thereof. The effector domain may recruit one or more protein domains that suppress the expression of a target gene, for example, through a scaffold protein. In some embodiments, the effector domain may recruit or interact with a scaffold protein domain that recruits a PRMT protein, an HDAC protein, a SETDB1 protein, or a NuRD protein domain.

[0108] In some embodiments, the effector domain includes a functional domain derived from zinc finger repressor proteins, such as the KRAB domain. The KRAB domain is found in approximately 400 human ZFP-based transcription factors. A description of the KRAB domain can be found, for example, in Ecco et al., Development (2017) 144(15):2719-29 and Lambert et al., Cell (2018) 172:650-65.

[0109] In a particular embodiment, the effector domain includes a repressor domain (e.g., KRAB) derived from KOX1 / ZNF10, KOX8 / ZNF708, ZNF43, ZNF184, ZNF91, HPF4, HTF10, or HTF34. In some embodiments, the effector domains are ZIM3, ZNF436, ZNF257, ZNF675, ZNF490, ZNF320, ZNF331, ZNF816, ZNF680, ZNF41, ZNF189, ZNF528, ZNF543, ZNF554, ZNF140, ZNF610, ZNF264, ZNF350, ZNF8, ZNF582, ZNF30, ZNF324, ZNF98, ZNF669, ZNF677, ZNF596, ZNF214, ZNF37, ZNF34, Z The system includes a repressor domain (e.g., KRAB) derived from NF250, ZNF547, ZNF273, ZNF354, ZFP82, ZNF224, ZNF33, ZNF45, ZNF175, ZNF595, ZNF184, ZNF419, ZFP28-1, ZFP28-2, ZNF18, ZNF213, ZNF394, ZFP1, ZFP14, ZNF416, ZNF557, ZNF566, ZNF729, ZIM2, ZNF254, ZNF764, ZNF785, or any combination thereof. For example, the repressor domain may be a KRAB domain derived from KOX1, ZIM3, ZFP28, or ZN627. In a particular embodiment, the repressor domain is a ZIM3 KRAB domain. In a further embodiment, the effector domain is derived from a human protein, such as human ZIM3, human KOX1, human ZFP28, or human ZN627.

[0110] Examples of effector domain sequences, or protein sequences containing them, that can reduce or silence target gene expression are provided in Table 4 below (SEQ: Sequence ID). Further examples of repressor and transcriptional repressor domains can be found, for example, in PCT Patent Application Publication No. 2021 / 226077 and Tycko et al., Cell (2020) 183(7):2020-35, each of which is incorporated herein by reference in its entirety. Table 4. Exemplary effector domains that can reduce or silence gene expression.

[0111] [Table 4] TIFF2026510339000015.tif248163TIFF2026510339000016.tif250164TIFF2026510339000017.tif250164 TIFF2026510339000018.tif250164TIFF2026510339000019.tif250164TIFF2026510339000020.tif111164

[0112] Functional analogs of any one of the proteins listed above, i.e., molecules that have the same or substantially the same biological function (e.g., 70% or more, 80% or more, 90% or more, 95% or more, or 98% or more of the protein's transcription factor function), are included in this disclosure. For example, a functional analog may be an isoform or variant of one of the proteins listed above, including, for example, a portion of the protein having or not having additional amino acid residues, and / or a mutation of the protein. In some embodiments, a functional analog has at least 75%, 80%, 85%, 90%, 95%, 98%, or 99% sequence identity to one of the sequences listed in Table 4. Homologs, orthologues, and variants of the proteins listed above are also intended.

[0113] In certain embodiments, the epigenetic editor described herein comprises a KRAB domain derived from KOX1, ZIM3, ZFP28, or ZN627, and / or an effector domain derived from KAP1, MECP2, HP1a, HP1b, CBX8, CDYL2, TOX, TOX3, TOX4, EED, EZH2, RBBP4, RCOR1, or SCML2, and the parent protein is optionally a human protein. In certain embodiments, the epigenetic editor described herein comprises a domain derived from KOX1, ZIM3, ZFP28, and / or ZN627, and the parent protein is optionally a human protein. In certain embodiments, the epigenetic editor may comprise a KRAB domain derived from KOX1 (ZNF10), for example, human KOX1. In certain embodiments, the epigenetic editor may comprise a KRAB domain derived from ZIM3 (ZNF657 or ZNF264), for example, human ZIM3. In certain embodiments, the epigenetic editor may include a KRAB domain derived from ZFP28, for example, human ZFP28. In certain embodiments, the epigenetic editor may include a KRAB domain derived from ZN627, for example, human ZN627. In certain embodiments, the epigenetic editor described herein may include a CDYL2, for example, human CDYL2 and / or a TOX domain (for example, a human TOX domain) in combination with a KOX1 KRAB domain (for example, a human KOX1 KRAB domain).

[0114] In certain embodiments, the epigenetic effector described herein includes a repressor domain (SEQ ID NO: 89) derived from KOX1 / ZNF10. For example, the repressor domain may include the sequence of SEQ ID NO: 89, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 89.

[0115] In a particular embodiment, the epigenetic effector described herein comprises a repressor domain derived from KOX1 / ZNF10, as shown in Table 5 below. Table 5. Exemplary effector domains derived from KOX1 / ZNF10

[0116] [Table 5]

[0117] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 565, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 565.

[0118] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 566, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 566.

[0119] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 567, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 567.

[0120] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 568, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 568.

[0121] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 569, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 569.

[0122] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 570, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 570.

[0123] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 571, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 571.

[0124] In certain embodiments, the repressor domain may include the amino acid sequence of SEQ ID NO: 572, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 572.

[0125] B. DNA methyltransferase In some embodiments, the effector domains of the epigenetic editors described herein alter target gene expression through DNA modification, such as methylation. Highly methylated regions (areas) of DNA tend to have lower transcriptional activity than less methylated regions. DNA methylation primarily occurs at CpG sites (abbreviations for "C-phosphate-G-" or "cytosine-phosphate-guanine" sites). Numerous mammalian genes have promoter regions near or containing CpG islands (nucleic acid regions with a high frequency of CpG dinucleotides).

[0126] The effector domains described herein may, for example, be DNA methyltransferases (DNMTs) or their catalytic domains, or may be capable of recruiting DNA methyltransferases. DNMTs include enzymes that catalyze the transfer of methyl groups to DNA nucleotides, such as classical cytosine-5 DNMTs (e.g., DNMT1, DNMT3A, DNMT3B, and DNMT3C) that catalyze the addition of methyl groups to genomic DNA. The term also includes non-classical family members that do not catalyze methylation themselves but recruit (including activate) catalytically active DNMTs, a non-limiting example of such DNMTs being DNMT3L. See, for example, Lyko, Nat Review (2018) 19:81-92. Unless otherwise indicated, a DNMT domain may refer to a polypeptide domain derived from a catalytically active DNMT (e.g., DNMT1, DNMT3A, and DNMT3B) or a catalytically inactive DNMT (e.g., DNMT3L). DNMTs can repress the expression of target genes through the recruitment of repressive regulatory proteins. In some embodiments, methylation occurs in CG (or CpG) dinucleotide sequences. In some embodiments, methylation occurs in CHG or CHH sequences, where H is one of A, T, or C.

[0127] In some embodiments, the DNMTs described herein may be animal DNMTs (e.g., mammalian DNMTs), plant DNMTs, fungal DNMTs, or bacterial DNMTs. Bacterial DNMT can be obtained from bacterial species (e.g., cocci, bacilli, spiral bacteria, or intracellular Gram-positive or Gram-negative bacteria). In certain embodiments, the bacterial species is Mycoplasmatales bacterium, Mycoplasma marinum, or Spiroplasma chinense. In certain embodiments, the bacterial species is M. penetrans, S. monvier, H. parainfluenzae, A. luteus, H. aegyptius, H. haemolyticus, Moraxella, E. coli, T. aquaticus, C. crescentus, or C. difficile. (difficile) not. In certain embodiments, the epigenetic editor described herein includes a DNMT domain containing sequence number 601, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 601. In certain embodiments, the epigenetic editor described herein includes sequence number 602, or at least identical to sequence number 602. The epigenetic editor described herein includes a DNMT domain containing a sequence that is 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 603.

[0128] In certain embodiments, the DNMT in the epigenetic editor described herein may include, for example, DNMT1, DNMT3A, DNMT3B, and / or DNMT3C. In some embodiments, the DNMT is a mammalian (e.g., human or mouse) DNMT. In certain embodiments, the DNMT is DNMT3A (e.g., human DNMT3A). In certain embodiments, the epigenetic editor described herein includes a DNMT3A domain containing sequence number 574, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 574. In certain embodiments, the epigenetic editor described herein includes a DNMT3A domain containing sequence number 575, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 575. In some embodiments, the DNMT3A domain may have mutations at, for example, position H739 (e.g., H739A or H739E), position R771 (e.g., R771L), and / or position R836 (e.g., R836A or R836Q), or any combination thereof (numbering follows sequence number 574).

[0129] In some embodiments, the effector domains described herein may be DNMT-like domains. As used herein, “DNMT-like domain” means a DNMT regulator that can activate or recruit other DNMT domains but does not possess methylation activity itself. In some embodiments, the DNMT-like domain is a mammalian (e.g., human or mouse) DNMT-like domain. In certain embodiments, the DNMT-like domain is DNMT3L, which may be, for example, human DNMT3L or mouse DNMT3L. In certain embodiments, the epigenetic editor described herein includes a DNMT3L domain containing sequence number 578, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 578. In a particular embodiment, the epigenetic editor described herein includes a DNMT3L domain containing a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 579. In a particular embodiment, the epigenetic editor described herein includes a DNMT3L domain containing a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 580. In certain embodiments, the epigenetic editor described herein includes a DNMT3L domain containing a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to sequence number 581. In some embodiments, the DNMT3L domain may have mutations corresponding to, for example, position D226 (e.g., D226V), position Q268 (e.g., Q268K), or both (numbering follows sequence number 578).

[0130] In certain embodiments, the epigenetic editor described herein may include both DNMT and DNMT-like effector domains. For example, the epigenetic editor may include a DNMT3A-3L domain, where DNMT3A and DNMT3L may be covalently linked. In other embodiments, the epigenetic editor described herein may include an effector domain containing only a DNMT3A domain (e.g., human DNMT3A) or only a DNMT-like domain (e.g., DNMT3L, which may be human or mouse DNMT3L).

[0131] Table 6 below provides exemplary DNMTs that may be part of the epigenetic effector domains described herein, or from which the effector domains of the epigenetic editors described herein may originate. Table 6. Exemplary DNMT sequences

[0132] [Table 6] TIFF2026510339000023.tif236158

[0133] Functional analogs of any one of the proteins listed above, i.e., molecules that have the same or substantially the same biological function (e.g., 70% or more, 80% or more, 90% or more, 95% or more, or 98% or more of the protein's DNA methylation or recruitment function), are included in this disclosure. For example, a functional analog may be an isoform or variant of the protein listed above, including, for example, a portion of the protein having or not having additional amino acid residues, and / or a mutation in the protein. In some embodiments, a functional analog has at least 75, 80, 85, 90, 95, 98, or 99% sequence identity to one of the sequences listed in Table 6. In some embodiments, the effector domain as defined herein includes only the functional domain (or functional analog thereof) of the protein listed above, e.g., a catalytic domain or a recruitment domain. In some embodiments, the effector domain as defined herein includes one or more epigenetic effector domains selected from Table 6, or their functional homologs, orthologues, or variants.

[0134] As used herein, a DNMT domain (e.g., a DNMT3A domain or a DNMT3L domain) refers to a protein domain that is identical to the parent protein (e.g., human or mouse DNMT3A or DNMT3L) or its functional analog (e.g., having a functional fragment of the parent protein, e.g., a catalytic fragment or a mobilizing fragment, and / or having a mutation that improves the activity of the DNMT protein).

[0135] The epigenetic editors described herein can induce methylation in, for example, one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, thirty, forty, fifty, sixty, seventy, eighteen, ninety, one hundred, two hundred, three hundred, four hundred, five hundred, six hundred, seventy, eighteen, ninety, one hundred, two hundred, three hundred, four hundred, five hundred, six hundred, seventy, eighteen, ninety, or one thousand or more CpG dinucleotide sequences within a target gene or chromosome. The CpG dinucleotide sequences may be located within or near the target gene in a CpG island, or they may be located in a region that is not a CpG island. A CpG island generally refers to a nucleic acid sequence or chromosomal region that contains CpG dinucleotides at a high frequency. For example, a CpG island may contain at least 50% GC content. A CpG island may have a high observed-to-expected CpG ratio, for example, at least 60%. As used herein, the observed-to-expected CpG ratio is determined by the number of CpGs × (sequence length) / (number of Cs × number of Gs). In some embodiments, a CpG island has an observed-to-expected CpG ratio of at least 60%, 70%, 80%, 90%, or higher. A CpG island may be, for example, a sequence or region of at least 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, or 800 nucleotides. In some embodiments, one or fewer than two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, thirty, forty, or fifteen CpG dinucleotides are methylated by an epigenetic editor.

[0136] In some embodiments, the epigenetic editors herein induce methylation in hypomethylated nucleic acid sequences, i.e., sequences in which a methyl group on the 5-methylcytosine nucleotide (e.g., in CpG) may be absent compared to a standard control. Hypomethylation may occur, for example, in senescent cells or cancer (e.g., in the early stages of tumorigenesis) compared to young cells or non-cancerous cells, respectively.

[0137] In some embodiments, the epigenetic editors described herein induce methylation in highly methylated nucleic acid sequences.

[0138] In some embodiments, methylation can be introduced by an epigenetic editor at sites other than CpG dinucleotides. For example, a target gene sequence may be methylated at the C nucleotide of a CpA, CpT, or CpC sequence. In some embodiments, the epigenetic editor comprises a DNMT3A domain and results in methylation at CpG, CpA, CpT, CpC sequences, or any combination thereof. In some embodiments, the epigenetic editor comprises a DNMT3A domain that lacks a regulatory subdomain and retains only the catalytic domain. In some embodiments, an epigenetic editor comprising a DNMT3A catalytic domain results in methylation only at CpG sequences. In some embodiments, an epigenetic editor comprising a DNMT3A domain containing a mutation, e.g., R836A or R836Q mutation (numbering according to Sequence ID No. 574), has higher methylation activity at CpA, CpC, and / or CpT sequences compared to an epigenetic editor comprising a wild-type DNMT3A domain.

[0139] C. Histone Modifiers In some embodiments, the effector domain of the epigenetic editor described herein mediates histone modifications. Histone modifications play structural and biochemical roles in gene transcription, for example, by forming or disrupting nucleosome structures that bind to histones and prevent gene transcription. Examples of histone modifications include acetylation, deacetylation, methylation, phosphorylation, ubiquitination, and SUMOylation at their N-terminus ("histone tail"). These modifications maintain or specifically alter chromatin structure, thereby regulating responses occurring on chromosomal DNA, such as gene expression, DNA replication, and DNA repair. Post-translational modifications of histones are epigenetic regulatory mechanisms and are considered essential for gene regulation in eukaryotic cells. Recent studies have revealed that chromatin remodeling factors, such as SWI / SNF, RSC, NURF, and NRD, which modify nucleosome structure to facilitate access of transcription factors to DNA, as well as histone acetyltransferases (HATs) and histone deacetylases (HDACs) that regulate the acetylation state of histones, act as important regulators.

[0140] In particular, the unstructured N-terminus of histones can be modified by acetylation, deacetylation, methylation, ubiquitination, phosphorylation, SUMOylation, ribosylation, citrullination, O-GlcN acylation, or crotonylation, or any combination thereof. For example, histone acetyltransferase (HAT) utilizes acetyl-CoA as a cofactor to catalyze the transfer of an acetyl group to the epsilon-amino group of the lysine side chain. This neutralizes the positive charge of lysine, weakening the interaction between histone and DNA, thereby opening the chromosome and allowing transcription factors to bind and initiate transcription. Acetylation of K14 and K9 lysines of histone H3 by histone acetyltransferase enzymes may be associated with transcriptional competence in humans. Lysine acetylation can directly or indirectly create binding sites for chromatin-modifying enzymes that regulate transcriptional activation. On the other hand, histone methylation of lysine 9 on histone H3 may be associated with heterochromatin, or transcriptionally silent chromatin.

[0141] In certain embodiments, the effector domain of the epigenetic editor described herein comprises a histone methyltransferase domain. The effector domain may comprise, for example, a DOT1L domain, a SET domain, a SUV39H1 domain, a G9a / EHMT2 protein domain, an EZH1 domain, an EZH2 domain, a SETDB1 domain, or any combination thereof. In certain embodiments, the effector domain comprises a histone-lysine-N-methyltransferase SETDB1 domain.

[0142] In some embodiments, the effector domain includes a histone deacetylase protein domain. In certain embodiments, the effector domain includes an HDAC family protein domain, such as HDAC1, HDAC3, HDAC5, HDAC7, or HDAC9 protein domain. In certain embodiments, the effector domain includes a nucleosome remodeling and deacetylase complex (NURD), which removes acetyl groups from histones.

[0143] D. Other effects pedal domain In some embodiments, the effector domain comprises a tripartite motif-containing protein (TRIM28, TIF1-beta, or KAP1). In certain embodiments, the effector domain comprises one or more KAP1 proteins. The KAP1 protein in the epigenetic editor herein may form a complex with one or more other effector domains of the epigenetic editor, or with one or more proteins involved in regulating gene expression in the cellular environment. For example, KAP1 may be recruited by the KRAB domain of a transcriptional repressor. The KAP1 protein domain may interact with or recruit one or more protein complexes that reduce or silence gene expression. In some embodiments, KAP1 interacts with or recruits histone deacetylase proteins, histone-lysine methyltransferase proteins, chromatin remodeling proteins, and / or heterochromatin proteins. For example, the KAP1 protein domain may interact with or recruit heterochromatin protein 1 (HP1), SETDB1, HDAC, and / or NuRD protein complex components. In some embodiments, the KAP1 protein domain interacts with or recruits ZFP90 protein (e.g., isoform 2 of ZFP90) and / or FOXP3 protein. An exemplary KAP1 amino acid sequence is shown in SEQ ID NO: 629.

[0144] In some embodiments, the effector domain includes a protein domain that interacts with or is recruited by one or more DNA epigenetic marks. For example, the effector domain may include a methyl CpG-binding protein 2 (MECP2) protein that interacts with methylated DNA nucleotides within a target gene (which may or may not be located in a CpG island of the target gene). The MECP2 protein domain in the epigenetic editor described herein may induce a condensed chromatin structure, thereby reducing or silencing the expression of the target gene. In some embodiments, the MECP2 protein domain in the epigenetic editor described herein may interact with a histone deacetylase (e.g., HDAC), thereby repressing or silencing the expression of the target gene. In some embodiments, the MECP2 protein domain in the epigenetic editor described herein may block access of transcription factors or transcription activators to a target sequence, thereby repressing or silencing the expression of the target gene. An exemplary MECP2 amino acid sequence is shown in SEQ ID NO: 630.

[0145] Effector domains of epigenetic editors described herein include, for example, chromoshadow domains, ubiquitin-2-like Rad60 SUMO-like (Rad60-SLD / SUMO) domains, chromatin organizing modifier (Chromo) domains, and Yaf2 / RYBP. C-terminal binding motif domain (YAF2_RYBP), CBX family C-terminal motif domain (CBX7_C), zinc finger C3HC4 type (RING finger) domain (ZF-C3HC4_2), cytochrome b5 domain (Cyt-b5), helix-loop-helix domain (HLH), helix-hairpin-helix motif domain (e.g., HHH_3), high mobility group box domain (HMG-box), basic leucine zipper domain (e.g., bZIP_1 or bZIP_2), Myb_DNA binding domain, homeodomain, MYM type zinc finger domain with FCS sequence domain (ZF-FCS), interferon regulator 2-binding protein zinc Finger domain (IRF-2BP1_2), SSX repressor domain (SSXRD), B-box zinc finger domain (ZF-B_box), CXXC zinc finger domain (ZF-CXXC), chromosome condensation regulator 1 domain (RCC1), SRC homology 3 domain (SH3_9), sterilous alpha motif domain (SAM_1), sterilous alpha motif domain (SAM_2), sterilous alpha motif / pointed domain (SAM_PNT), Vestigial / Tondu family domain (Vg_Tdu), LIM domain, RNA recognition motif domain (RRM_1), pairing amphipathic helix domain (PAH), proteasome ATPase OB C-terminal domain (Prot_ATP_ID_OB), nervy homology 2 domain (NHR2), cleavage stimulator subunit 2 hinge domain (CSTF2_hinge), PPAR gamma N-terminal region domain (PPARgamma_N), CDC48N-terminal domain (CDC48_2), WD40 repeat domain (WD40), Fip1 motif domain (Fip1), PDZ domain (PDZ_6), von Willebrand factor type C domain (VWC), NAB conserved region 1 domain (NCD1), S1 RNA binding domain (S1), HNF3 C-terminal domain (HNF_C), Tudor domain (Tudor_2), histone-like transcription factor (CBF / NF-Y) and archaeal histone domain (CBFD_NFYB_HMF), zinc finger protein domain (DUF3669), EGF-like domain (cEGF), GATA zinc finger domain (GATA), TEA / ATTS domain (TEA), phorbol ester / diacylglycerol-binding domain (C1-1), Polycomb-like MTF2 factor 2 domain (Mtf2_C), FOXO protein family transactivation domain (FOXO-TAD), homeobox KN domain (Homeobox_KN), BED zinc finger domain (ZF-BED), C3HC4-type RING domain zinc finger (ZF-C3HC4_4), RAD51 interaction motif domain (RAD51_interact), methyl (nethyl)-CpG-binding domain protein MBD p55 binding region (MBDa), notch domain, Raf-like Ras binding domain (RBD), Spin / Ssty family domain (Spin-Ssty), PHD finger domain (PHD_3), low-density lipoprotein receptor domain class A (Ldl_recept_a), CS domain, DM DNA binding domain, and QLQ domain are also intended.

[0146] In some embodiments, the effector domain is a protein domain comprising the YAF2_RYBP domain, its homeodomain, or any combination thereof. In certain embodiments, the homeodomain of the YAF2_RYBP domain is a PRD domain, an NKL domain, a HOXL domain, or a LIM domain. In certain embodiments, the YAF2_RYBP domain may comprise a 32-amino acid Yaf2 / RYBP C-terminal binding motif domain (32 aa RYBP).

[0147] In some embodiments, the effector domain includes a protein domain selected from the group consisting of a SUMO3 domain, a chromodomain derived from M-phase phosphate protein 8 (MPP8), a chromoshadow domain derived from chromoboc1 (CBX1), and a SAM_1 / SPM domain derived from Scm Polycomb group protein homolog 1 (SCMH1).

[0148] In some embodiments, the effector domain includes an HNF3 C-terminal domain (HNF_C). The HNF_C domain may originate from FOXA1 or FOXA2. In certain embodiments, the HNF_C domain includes an EH1 (engrailed homology 1) motif.

[0149] In some embodiments, the effector domain may include an interferon regulator 2 binding protein zinc finger domain (IRF-2BP1_2), a Cyt-b5 domain derived from the DNA repair factor HERC2 E3 ligase, a variant SH3 domain (SH3_9) derived from bridging integrator 1 (BIN1), an HMG box domain derived from the transcription factor TOX, or a ZF-C3HC4_2 ring finger domain derived from the Polycomb component PCGF2, a chromodomain-helicase-DNA binding protein 3 (CHD3) domain, or a ZNF783 domain.

[0150] IV. Epigenetic Editor For example, epigenetic editors (i.e., epigenetic editing systems) are provided herein that use any combination of one or more DNA-binding domains and one or more effector domains (e.g., epigenetic repressor domains) described herein to direct epigenetic modifications (or multiple modifications) to a target sequence in a gene of interest. The DNA-binding domains (working in cooperation with guide polynucleotides, e.g., those described herein, where the DNA-binding domains are DNA-binding domains guided to polynucleotides) direct the effector domains to epigenetically modify the target sequence, resulting in persistent gene repression or silencing that can be inherited across generations of cells. In some embodiments, the epigenetic editors described herein can reversibly or irreversibly repress or silence genes in cells.

[0151] In certain embodiments, the epigenetic editors described herein comprise one or more fusion proteins, each comprising (1) a DNA-binding domain(s) and (2) an effector domain(s). The effector domains may reside on one or more fusion proteins comprised by the epigenetic editor. For example, a single fusion protein may comprise all effector domains having DNA-binding domains. Alternatively, the effector domains or a subset thereof may reside on separate fusion proteins, each having a DNA-binding domain (which may be the same or different). The fusion proteins described herein may further comprise one or more linkers (e.g., peptide linkers), a detectable tag, a nuclear localization signal (NLS), or any combination thereof. As used herein, “fusion protein” refers to a chimeric protein in which two or more coding sequences (e.g., of DNA-binding domains(s) and / or effector domains(s)) are directly or indirectly linked by covalent or non-covalent bonds.

[0152] In some embodiments, the epigenetic editors described herein include two, three, four, five, six, seven, eight, nine, ten, or more effector (e.g., repressor / repressor) domains, which may be identical or different. In certain embodiments, two or more of the effector domains function synergistically. Combinations of effector domains may include DNA methylation domains, histone deacetylation domains, histone methylation domains, and / or scaffold domains that recruit any of the above. For example, the epigenetic editors described herein may include one or more transcriptional repressor domains (e.g., KRAB domains, e.g., KOX1, ZIM3, ZFP28, or ZN627 KRAB) in combination with one or more DNA methylation domains (e.g., DNMT domains) and / or recruiter domains (e.g., DNMT3L domains). Such epigenetic editors may include, for example, the KRAB domain, the DNMT3A domain, and the DNMT3L domain. In some embodiments, the epigenetic editor further includes additional effector domains (e.g., the KAP1, MECP2, HP1b, CBX8, CDYL2, TOX, TOX3, TOX4, EED, RBBP4, RCOR1, or SCML2 domains). In some embodiments, the additional effector domains are the CDYL2, TOX, TOX3, TOX4, or HP1a domains. For example, the epigenetic editors described herein may include the CDYL2 and / or TOX domains in combination with the KRAB domain (e.g., the KOX1 KRAB domain).

[0153] A. Linker The fusion proteins described herein may include one or more linkers connecting the components of the epigenetic editor. The linkers may be peptide linkers or non-peptide linkers.

[0154] In some embodiments, one or more linkers used in the epigenetic editor provided herein are peptide linkers, i.e., linkers containing a peptide portion. The peptide linker may be of any length applicable to the epigenetic editor fusion protein described herein. In some embodiments, the linker may contain a peptide of 1 to 200 (e.g., 1 to 80) amino acids. In the half embodiment, the linkers are 1-5, 1-10, 1-20, 1-30, 1-40, 1-50, 1-60, 1-80, 1-100, 1-150, 1-200, 5-10, 5-20, 5-30, 5-40, 5-60, 5-80, 5-100, 5-150, 5-200, 10-20, 10-30, 10-40, 10-50, 10-60, 10-80, 10-100, 10-150, 10-200, 20-30, 20-40, 20-50, 20-60, 20-80, 20-1 Includes lengths of 00, 20-150, 20-200, 30-40, 30-50, 30-60, 30-80, 30-100, 30-150, 30-200, 40-50, 40-60, 40-80, 40-100, 40-150, 40-200, 50-60, 50-80, 50-100, 50-150, 50-200, 60-80, 60-100, 60-150, 60-200, 80-100, 80-150, 80-200, 100-150, 100-200, or 150-200 amino acids. Longer or shorter linkers are also intended. In some embodiments, the peptide linker is the length of 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 25, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100 amino acids. For example, the peptide linker may be the length of 4, 5, 16, 20, 24, 27, 32, 40, 64, 92, or 104 amino acids. The peptide linker may be a movable linker or a rigid linker.In certain embodiments, the peptide linker includes one amino acid sequence from sequence numbers 631-637 and 664-665, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto.

[0155] In certain embodiments, the peptide linker is an XTEN linker. Such a linker may contain a portion of an XTEN sequence, which is an unstructured hydrophilic polypeptide consisting only of residues G, S, P, T, E, and A (Schellenberger et al., Nat Biotechnol (2009) 27(1):1186-90). As used herein, the term "XTEN" refers to a recombinant peptide or polypeptide lacking hydrophobic amino acid residues. XTEN linkers are typically unstructured and contain a limited set of native amino acids. Fusion of XTEN to a protein alters its hydrodynamic properties and reduces the clearance and degradation rates of the fusion protein. These XTEN fusion proteins are produced using recombinant technology without requiring chemical modification and are degraded by the native pathway. XTEN linkers may be, for example, 5, 10, 16, 20, 26, or 80 amino acid lengths. In some embodiments, the XTEN linker is 16 amino acids long. In some embodiments, the XTEN linker is 80 amino acids long. In certain embodiments, the XTEN linker may be XTEN10, XTEN16, XTEN20, or XTEN80. In certain embodiments, the XTEN linker may contain any one amino acid sequence from SEQ ID NOs. 638 to 643, or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to it. In certain embodiments, the XTEN linker contains the amino acid sequence of SEQ ID NO. 638. In certain embodiments, the XTEN linker contains the amino acid sequence of SEQ ID NO. 643.

[0156] In some embodiments, one or more linkers used in the epigenetic editor provided herein are non-peptide linkers. For example, the linker may be a carbon bond, a disulfide bond, or a carbon-heteroatom bond. In certain embodiments, the linker is an amide-linked carbon-nitrogen bond. In certain embodiments, the linker may be a cyclic or acyclic, substituted or unsubstituted, or branched or unbranched aliphatic or heteroaliphatic linker.

[0157] In some embodiments, one or more linkers used in the epigenetic editor provided herein are polymers (e.g., polyethylene, polyethylene glycol, polyamide, polyester, etc.). The linker may include, for example, monomers, dimers, or polymers of aminoalkanoic acids, aminoalkanoic acids (e.g., glycine, ethaneic acid, alanine, beta-alanine, 3-aminopropanoic acid, 4-aminobutanoic acid, 5-pentanoic acid, etc.), monomers, dimers, or polymers of aminohexanoic acid (Ahx), or polyethylene glycol moieties (PEG), or aryl or heteroaryl moieties. In certain embodiments, the linker may be based on a carbocyclic moiety (e.g., cyclopentane or cyclohexane) or a phenyl ring. The linker may include a functionalized moiety that facilitates the bonding of a nucleophile (e.g., thiol, amino) peptide to the linker. Any electrophile may be used as a part of the linker. Exemplary electrophiles include, but are not limited to, activated esters, activated amides, alkyl halides, aryl halides, acyl halides, and isothiocyanates.

[0158] Various linker lengths and mobilities can be utilized between any two components of the epigenetic editor (e.g., between an effector domain (e.g., a repressor domain) and a DNA-binding domain (e.g., a Cas9 domain), or between a first effector domain and a second effector domain). The linker may range from highly mobile linkers, e.g., glycine / serine-rich linkers, to more rigid linkers in order to achieve the optimal length for effector domain activity for a given application. In some embodiments, the more mobile linker is a glycine / serine-rich linker (GS-rich linker) in which more than 45% (e.g., more than 48%, more than 50%, more than 55%, more than 60%, more than 70%, more than 80%, or more than 90%) of the residues are glycine or serine residues. Non-limiting examples of GS-rich linkers include (GGGGS)n (SEQ ID NO: 664), (G)n, and W linkers (SEQ ID NO: 637). In some embodiments, more rigid linkers are of the form (EAAAK)n (SEQ ID NO: 665), (SGGS)n (SEQ ID NO: 631), and (XP)n). In the above formulas for movable and rigid linkers, n can be any integer from 1 to 30. In some embodiments, n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15. In some embodiments, the linker includes a (GGS)n motif, where n is 1, 3, or 7. In some embodiments, the linker includes a (GGGGS)n motif, where n is 4 (SEQ ID NO: 636).

[0159] In some embodiments, the linker in the epigenetic editor described herein includes, for example, a nuclear localization signal having one of the amino acid sequences of SEQ ID NOs.644-649. In some embodiments, the linker in the epigenetic editor described herein includes an expression tag, such as a detectable tag, such as a green fluorescent protein.

[0160] B. Nuclear localization signals The fusion proteins described herein may contain one or more nuclear localization signals, and in certain embodiments, two or more nuclear localization signals. For example, a fusion protein may contain one, two, three, four, five, six, seven, eight, nine, ten, or more nuclear localization signals. As used herein, a “nuclear localization signal” (NLS) is an amino acid sequence that directs a protein toward the nucleus. In certain embodiments, the NLS may be an SV40 NLS (for example, having the amino acid sequence of SEQ ID NO: 644). The fusion protein may contain an NLS at its N-terminus, C-terminus, or both, and / or the NLS may be embedded in the middle of the fusion protein (for example, at the N-terminus or C-terminus of a DNA-binding domain or effector domain).

[0161] In some embodiments, the fusion protein may contain two NLSs. The fusion protein may contain two NLSs at its N-terminus or C-terminus. The fusion protein may contain one NLS located at its N-terminus and one NLS embedded in the middle of the fusion protein, or one NLS located at its C-terminus and one NLS embedded in the middle of the fusion protein. The fusion protein may contain two NLSs embedded in the middle of the fusion protein.

[0162] In some embodiments, the fusion protein may contain four NLSs. The fusion protein may contain at least two (e.g., two, three, or four) NLSs at its N-terminus or C-terminus. The fusion protein may contain at least one (e.g., one, two, three, or four) NLSs embedded in the middle of the fusion protein. In certain embodiments, the fusion protein may contain two NLSs at its N-terminus and two NLSs at its C-terminus.

[0163] The NLSs described herein may be endogenous NLS sequences. In certain embodiments, the NLSs described herein include a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to any one of the amino acid sequences of SEQ ID NOs. 644, 649, or a selected sequence. In certain embodiments, the NLS includes the amino acid sequence of SEQ ID NO. 644. Additional NLSs are known in the art.

[0164] In some embodiments, an epigenetic editor comprising a fusion protein having at least one NLS at the N-terminus and at least one NLS at the C-terminus can increase the efficiency of the epigenetic editor by at least 5%, at least 10%, at least 15%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 100%, at least 200%, at least 300%, at least 400%, at least 500%, at least 600%, at least 700%, at least 80%, at least 90%, at least 100%, at least 200%, at least 300%, at least 400%, at least 500%, at least 600%, at least 700%, at least 800%, at least 900%, at least 1,000%, at least 5,000%, at least 10,000%, at least 50,000%, at least 100,000%, or more compared to an epigenetic editor comprising a corresponding fusion protein not having at least one NLS at the N-terminus and at least one NLS at the C-terminus.

[0165] In some embodiments, an epigenetic editor comprising a fusion protein having two NLS at the N-terminus and two NLS at the C-terminus can increase the efficiency of the epigenetic editor by at least 5%, at least 10%, at least 15%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 100%, at least 200%, at least 300%, at least 400%, at least 500%, at least 600%, at least 700%, at least 800%, at least 90%, at least 100%, at least 200%, at least 300%, at least 400%, at least 500%, at least 600%, at least 700%, at least 800%, at least 900%, at least 1,000%, at least 5,000%, at least 10,000%, at least 50,000%, at least 100,000%, or more compared to an epigenetic editor comprising a corresponding fusion protein not having two NLS at the N-terminus and two NLS at the C-terminus.

[0166] C. Tag The epigenetic editors provided herein may include one or more additional sequences (tags) for tracking, detection, and localization of the editor. In some embodiments, the epigenetic editor includes one, two, three, four, five, six, seven, eight, nine, ten, or more detectable tags. Each of the detectable tags may be the same or different.

[0167] For example, an epigenetic editor fusion protein may include a cytoplasmic localization sequence, an external transport sequence, such as a nuclear export sequence, or other localization sequences, as well as sequence tags useful for solubilization, purification, or detection of the fusion protein. Suitable protein tags provided herein include, but are not limited to, biotin carboxylase carrier protein (BCCP) tags, myc tags, calmodulin tags, FLAG tags, hemagglutinin (HA) tags, polyhistidine tags (also known as histidine tags or His tags), maltose-binding protein (MBP) tags, nus tags, glutathione-S-transferase (GST) tags, green fluorescent protein (GFP) tags, thioredoxin tags, S tags, Softag (e.g., Softag 1 or Softag 3), strep tags, biotin ligase tags, FlAsH tags, V5 tags, and SBP tags. Additional suitable sequences will be apparent to those skilled in the art.

[0168] D. Composition of fusion proteins The epigenetic editor fusion proteins described herein may have components structured in different configurations. For example, the DNA-binding domain may be C-terminus, N-terminus, or between two or more epigenetic effector domains or additional domains. In some embodiments, the DNA-binding domain is at the C-terminus of the epigenetic editor. In some embodiments, the DNA-binding domain is at the N-terminus of the epigenetic editor. In some embodiments, the DNA-binding domain is linked to one or more nuclear localization signals. In some embodiments, the DNA-binding domain is flanked on both sides by epigenetic effector domains or additional domains. In some embodiments, where "DBD" indicates the DNA-binding domain and "ED" indicates the effector domain, the epigenetic editor is: -N']-[ED1]-[DBD]-[ED2]-[C' -N']-[ED1]-[DBD]-[ED2]-[ED3]-[C' -N']-[ED1]-[ED2]-[DBD]-[ED3]-[C' or -N']-[ED1]-[ED2]-DBD]-[ED3]-[ED4]-[C' This includes the configuration.

[0169] In some embodiments, the epigenetic editor comprises a DNA-binding domain (DBD), a DNA methyltransferase (DNMT) domain, and a transcriptional repressor ("repressor") domain that represses or silences the expression of a target gene. The DBD, DNMT, and transcriptional repressor domains may be any combination described herein. The DBD, DNMT, and repressor domains may be in any configuration having any of the domains at the N-terminus, C-terminus, or intermediate of the fusion protein. In some embodiments, the epigenetic editor is N']-[DNMT domain]-[DBD]-[Repressor domain]-[C' N']-[Repressor Domain]-[DBD]-[DNMT Domain]-[C' N']-[DNMT domain]-[Repressor domain]-[DBD]-[C' or N']-[Repressor Domain]-[DNMT Domain]-[DBD]-[C' It contains a fusion protein having the following configuration.

[0170] In some embodiments, the linkage structure ]-[ in any one of the epigenetic editor structures is a linker, e.g., a peptide linker, a detectable tag, a peptide bond, a nuclear localization signal, and / or a promoter or regulatory sequence. In an epigenetic editor structure, multiple linkage structures ]-[ may be the same, or each may be a different linker, tag, NLS, or peptide bond. In some embodiments, the DNMT domain may include any one of the domains in Table 6, or any combination or homolog thereof. In certain embodiments, the DNMT domain includes DNMT3A or its truncated form, DNMT3L or its truncated form, or both. In certain embodiments, the DBD is a DNA-binding domain (e.g., dCas9) or ZFP domain guided to a catalytically inactive polynucleotide. In certain embodiments, the repressor domain includes any one of the domains shown in Table 4 or 5, or any combination or homolog thereof. For example, the repressor domain may be a KRAB domain. In certain embodiments, the repressor domain is the ZFP28, ZN627, KAP1, MeCP2, HP1b, CBX8, CDYL2, TOX, Tox3, Tox4, EED, RBBP4, RCOR1, or SCML2 domain, or a fusion of two of the above domains (for example, a fusion of the N-terminal and C-terminal regions of ZIM3 and KOX1 KRAB). In certain embodiments, the repressor domain is a KRAB domain derived from ZFP28, ZN627, ZIM3, or KOX1.

[0171] In some embodiments, the epigenetic editor is N']-[DNMT3A-DNMT3L]-[DBD]-[Repressor]-[C' N']-[Repressor]-[DBD]-[DNMT3A-DNMT3L]-[C' N']-[Repressor]-[DBD]-[DNMT3A]-[C' N']-[DNMT3A]-[DBD]-[Repressor]-[C' N']-[Repressor]-[DBD]-[DNMT3A]-[DNMT3L]-[C' N']-[DNMT3A]-[DNMT3L]-[DBD]-[Repressor]-[C' N']-[DNMT3A]-[DBD]-[C' N']-[DBD]-[DNMT3A]-[C' N']-[DNMT3L]-[DBD]-[C' N']-[DBD]-[DNMT3L]-[C' The configuration includes a selection from, where [DNMT3A-DNMT3L] indicates that the DNMT3A and DNMT3L domains are directly fused via a peptide bond, and [DNMT3A-DNMT3L] is any one of the linkers described herein, a detectable tag, an affinity domain, a peptide bond, a nuclear localization signal, a promoter, and / or a regulatory sequence. The DBD, repressor, DNMT3A, and DNMT3L domains may be any combination of those described herein. For example, the DNMT3A and DNMT3L domains may be selected from those in Table 6. In certain embodiments, the DBD is a CRISPR-associated protein domain (e.g., dCas9) or a ZFP domain, the repressor domain is a KRAB domain derived from KOX1, ZIM3, ZFP28, or ZN627, the DNMT3A domain is a human DNMT3A domain, and the DNMT3L domain is a human or mouse DNMT3L domain, and any combination of these components is also contemplated by this disclosure.

[0172] In some embodiments, the epigenetic editor is N']-[DNMT3A]-[DBD]-[SETDB1]-[C' N']-[DNMT3A]-[DNMT3L]-[DBD]-[SETDB1]-[C' N']-[DNMT3A-DNMT3L]-[DBD]-[SETDB1]-[C' N']-[SETDB1]-[DBD]-[DNMT3A]-[DNMT3L]-[C' N']-[SETDB1]-[DBD]-[DNMT3A]-[C' Includes configurations selected from, Here, [DNMT3A-DNMT3L] indicates that the DNMT3A and DNMT3L domains are directly fused via a peptide bond, and the linking structure [-] is any one of the linkers described herein, a detectable tag, an affinity domain, a peptide bond, a nuclear localization signal, a promoter, and / or a regulatory sequence. The DBD, SETDB1, DNMT3A, and DNMT3L domains may be any combination described herein. In certain embodiments, the DBD is a CRISPR-associated protein domain (e.g., dCas9) or a ZFP domain, the SETDB1 domain is derived from human SETDB1, ZIM3, ZFP28, or ZN627, the DNMT3A domain is a human DNMT3A domain, and the DNMT3L domain is a human or mouse DNMT3L domain, and any combination of these components is also contemplated by this disclosure.

[0173] The specific constructs intended herein include: DNMT3A-DNMT3L-XTEN80-NLS-dCas9-NLS-XTEN16-KOX1 KRAB (configuration 1), DNMT3A-DNMT3L-XTEN80-NLS-ZFP domain-NLS-XTEN16-KOX1 KRAB (Configuration 2), NLS-DNMT3A-DNMT3L-XTEN80-dCas9-XTEN16-KOX1 KRAB-NLS (configuration 3), NLS-DNMT3A-DNMT3L-XTEN80-ZFP domain-XTEN16-KOX1 KRAB-NLS (Configuration 4) NLS-NLS-DNMT3A-DNMT3L-XTEN80-dCas9-XTEN16-KOX1 KRAB-NLS-NLS (configuration 5), and NLS-NLS-DNMT3A-DNMT3L-XTEN80-ZFP domain-XTEN16-KOX1 KRAB-NLS-NLS (Configuration 6) These are some examples. DNMT3L and DNMT3A may be derived from human parent proteins, mouse parent proteins, or any combination thereof. In certain embodiments, DNMT3L and DNMT3A are derived from mouse and human parent proteins (mDNMT3L and hDNMT3A), respectively. In certain embodiments, both DNMT3L and DNMT3A are derived from human parent proteins (hDNMT3L and hDNMT3A). In some embodiments, dCas9 is dSpCas9. In some embodiments, KOX1 is human KOX1. Any of configurations 1-6 in which the KOX1 KRAB domain is replaced by a ZFP28, ZN627, or ZIM3 KRAB domain is also intended. In some embodiments, ZFP28, ZN627, and ZIM3 are human ZFP28, ZN627, and ZIM3, respectively. In certain embodiments, the fusion construct is: NLS-NLS-hDNMT3A-hDNMT3L-XTEN80-dCas9-XTEN16-KOX1 KRAB-NLS-NLS (configuration 7), NLS-NLS-DNMT3A-DNMT3L-XTEN80-ZFP domain-XTEN16-KOX1 KRAB-NLS-NLS (configuration 8), NLS-NLS-hDNMT3A-hDNMT3L-XTEN80-dCas9-XTEN16-ZFP28 KRAB-NLS-NLS (configuration 9), NLS-NLS-DNMT3A-DNMT3L-XTEN80-ZFP domain-XTEN16-ZFP28 KRAB-NLS-NLS (configuration 10), NLS-NLS-hDNMT3A-hDNMT3L-XTEN80-dCas9-XTEN16-ZN627 KRAB-NLS-NLS (configuration 11), NLS-NLS-DNMT3A-DNMT3L-XTEN80-ZFP domain-XTEN16-ZN627 KRAB-NLS-NLS (configuration 12), NLS-NLS-hDNMT3A-hDNMT3L-XTEN80-dCas9-XTEN16-ZIM3 KRAB-NLS-NLS (configuration 13), or NLS-NLS-DNMT3A-DNMT3L-XTEN80-ZFP domain-XTEN16-ZIM3 KRAB-NLS-NLS (Configuration 14) It may have.

[0174] In certain embodiments, the fusion constructs described herein may have configuration 1 and include sequences that are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto. In the following sequence number 658, the XTEN linker is shown in underline, the W linker in bold, underline, and italics, the NLS sequence in bold, the DNMT3A sequence in italics, the DNMT3L sequence in underline and italics, the dCas9 domain in bold and italics, and the KOX1 KRAB domain in underline and bold. TIFF2026510339000024.tif182153

[0175] In certain embodiments, the fusion constructs described herein may include the sequence provided below (SEQ ID NO: 1495), or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto. In SEQ ID NO: 1495 below, the XTEN linker is underlined, the W linker is bold, underlined, and italicized, the NLS sequence is bold, the DNMT3A sequence is italicized, the DNMT3L sequence is underlined and italicized, the dCas9 domain is bold and italicized, and the KOX1 KRAB domain is underlined and bold. TIFF2026510339000025.tif182153

[0176] In certain embodiments, the fusion constructs described herein may have configuration 2 and include sequences that are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto. In the following sequence number 659, the XTEN linker is shown underlined, the W linker is shown in bold, underlined, and italicized, the NLS sequence is shown in bold and underlined, the DNMT3A sequence is shown in italicized, the DNMT3L sequence is shown underlined and italicized, the ZFP domain is shown in bold, and the KOX1 KRAB domain is shown underlined and in bold. The variable amino acid represented by X is an amino acid of the DNA recognition helix of the zinc finger, and the italicized XX may be TR, LR, or LK. TIFF2026510339000026.tif86154 In certain embodiments, the six "XXXXXXX" regions in SEQ ID NO: 659 contain, in order, the F1-F6 amino acid sequences shown in Table 1 for any one of ZF001-ZF048. [Linker] represents a linker sequence. In some embodiments, one or both linker sequences may be TGSQKP (SEQ ID NO: 651). In some embodiments, one or both linker sequences may be TGGGGSQKP (SEQ ID NO: 652). In some embodiments, one linker sequence may have the amino acid sequence of SEQ ID NO: 651, and the other linker sequence may have the amino acid sequence of SEQ ID NO: 652.

[0177] In certain embodiments, the fusion constructs described herein may include the sequence provided below (SEQ ID NO: 1496), or a sequence that is at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto. In SEQ ID NO: 1496 below, the XTEN linker is underlined, the W linker is in bold, underlined, and italicized, the NLS sequence is in bold and underlined, the DNMT3A sequence is in italicized, the DNMT3L sequence is underlined and italicized, the ZFP domain is in bold, and the KOX1 KRAB domain is underlined and in bold. The variable amino acid represented by X is an amino acid of the DNA recognition helix of the zinc finger, and the italicized XX may be TR, LR, or LK. TIFF2026510339000027.tif86154

[0178] In a particular embodiment, the six "XXXXXXX" regions in SEQ ID NO: 1496 contain, in order, the F1-F6 amino acid sequences shown in Table 1 for any one of ZF001-ZF048. [Linker] represents a linker sequence. In some embodiments, one or both linker sequences may be TGSQKP (SEQ ID NO: 651). In some embodiments, one or both linker sequences may be TGGGGSQKP (SEQ ID NO: 652). In some embodiments, one linker sequence may have the amino acid sequence of SEQ ID NO: 651, and the other linker sequence may have the amino acid sequence of SEQ ID NO: 652.

[0179] In some embodiments, the fusion protein may further include a Dnmt3A ADD domain downstream of the Dnmt3A domain sequence disclosed, for example, in SEQ ID NOs. 658, 659, 1495, or 1496 disclosed above. In some embodiments, the ADD sequence is located between the Dnmt3A and Dnmt3L sequences of the fusion protein. In some embodiments, the ADD sequence is at the C-terminus of the Dnmt3A domain. In some embodiments, the Dnmt3A and ADD sequences are separated by a linker, for example, a linker disclosed herein. In some embodiments, the ADD sequence and the Dnmt3L sequence are separated by a linker, for example, a linker disclosed herein. In some embodiments, the ADD domain is part of the sequence: Includes TIFF2026510339000028.tif41155.

[0180] In certain embodiments, the fusion constructs described herein may have configuration 7 and may include sequences that are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto.

[0181] In certain embodiments, the fusion constructs described herein may have configuration 9 and may include sequences that are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto.

[0182] In certain embodiments, the fusion constructs described herein may have configuration 11 and may include sequences that are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto.

[0183] In certain embodiments, the fusion constructs described herein may have configuration 13 and may include sequences that are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical thereto.

[0184] In some embodiments, the fusion construct described herein (e.g., any one of constructs 1 to 14) is located within an expression construct that includes a WPRE sequence, a polyadenylation site, or both. In certain embodiments, the WPRE sequence is located in the 3' non-coding region. In certain embodiments, the WPRE sequence is located upstream of the polyadenylation site. In certain embodiments, the expression construct includes a fusion construct (e.g., any one of constructs 1 to 14) and a WPRE sequence located in the 3' non-coding region upstream of the polyadenylation site.

[0185] In some embodiments, the fusion constructs described herein may have one sequence of any one of the fusion proteins 1 to 12 shown in Example 12.

[0186] Multiple fusion proteins can be used to induce activation or repression of a target gene or multiple target genes. For example, an epigenetic editor fusion protein containing a DNA-binding domain (e.g., a dCas9 domain) and an effector domain may be co-delivered with two or more guide polynucleotides (e.g., gRNAs), each targeting a different target DNA sequence. The two target sites of the DNA-binding domain may be the same, located near each other, or separated by, for example, about 100 base pairs, 200 base pairs, 300 base pairs, 400 base pairs, 500 base pairs, 600 base pairs, or more. In addition, when targeting double-stranded DNA, for example, an endogenous locus, the guide polynucleotides may target the same strand or different strands (one or more positive strands and / or one or more negative strands).

[0187] V. Target sequence The epigenetic editors described herein may be directed to target sequences in the PCSK9 gene to induce epigenetic modification of the PCSK9 gene. Where used herein, “target sequence,” “target site,” or “target region” is a nucleic acid sequence present in the gene of interest, and in some cases, the target sequence may be outside but near the gene of interest, where gene expression is suppressed by methylation or binding of the target sequence by a repressor. In some embodiments, the target sequence may be a hypomethylated or hypermethylated nucleic acid sequence.

[0188] The target sequence can be located in any part of the target gene. In some embodiments, the target sequence is part of or near a non-coding sequence of the gene. In some embodiments, the target sequence is part of an exon of the gene. In some embodiments, the target sequence is part of or near a transcriptional regulatory sequence of the gene, such as a promoter or enhancer. In some embodiments, the target sequence is adjacent to, overlaps with, or encompasses a CpG island. In certain embodiments, the target sequence is located within approximately 3000, 2900, 2800, 2700, 2600, 2500, 2400, 2300, 2200, 2100, 2000, 1900, 1800, 1700, 1600, 1500, 1400, 1300, 1200, 1100, 1000, 900, 800, 700, 600, 500, 400, 300, 200, or 100 base pairs (bp) adjacent to the PCSK9 TSS. In certain embodiments, the target sequence is located within 500 bp adjacent to the PCSK9 TSS. In certain embodiments, the target sequence is located within 1000 bp adjacent to the PCSK9 TSS.

[0189] In some embodiments, the target sequence may hybridize to a guide polynucleotide sequence (e.g., gRNA) that forms a complex with a fusion protein containing a polynucleotide-guided DNA-binding domain (e.g., a CRISPR protein, e.g., dCas9) and one or more effector domains. The guide polynucleotide sequence may be designed to have complementarity with the target sequence or identity with respect to the opposing strand of the target sequence. In some embodiments, the guide polynucleotide includes a spacer sequence that is approximately 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to a protospacer sequence in the target sequence. In certain embodiments, the guide polynucleotide includes a spacer sequence that is 100% identical to a protospacer sequence in the target sequence.

[0190] In some embodiments, if the DNA-binding domain of the epigenetic editor described herein is a zinc finger array, the target sequence can be recognized by the zinc finger array.

[0191] In some embodiments, if the DNA-binding domain of the epigenetic editor described herein is a TALE, the target sequence can be recognized by the TALE.

[0192] The target sequences described herein may be specific to one copy of the target gene or to one allele of the target gene. Therefore, epigenetic modification and regulation of its expression may be specific to one copy or one allele of the target gene. For example, an epigenetic editor may suppress the expression of a particular copy (e.g., a copy associated with a disease or condition, or a copy with a mutation associated with a disease or condition) that has a target sequence recognized by a DNA-binding domain.

[0193] In some embodiments, the target PCSK9 genomic region may be contained within the following sequence (chr1:55038548-55040548), which may or may not have A at the terminal: TIFF2026510339000029.tif153155TIFF2026510339000030.tif31155

[0194] In some embodiments, the target sequence may be GRCh38 Chr1:55039228-55040296 as shown below: TIFF2026510339000031.tif100155

[0195] VI. Epigenetic Modification The epigenetic editors described herein can perform sequence-specific epigenetic modifications (e.g., alterations of chemical modifications) of a target gene having a target sequence. Such epigenetic regulation may be safer and more easily reversible than regulation resulting from gene editing, for example, the generation of DNA double-strand breaks. In some embodiments, the epigenetic regulation may reduce or silence the target gene. In some embodiments, the modification is at a specific site in the target sequence. In some embodiments, the modification is at a specific allele of the target gene. Thus, the epigenetic modification results in regulation (e.g., reduction) of the expression of one copy of the target gene having a specific allele, but not necessarily affecting other copies of the target gene. In some embodiments, the specific allele is associated with a disease, condition, or disorder.

[0196] In some embodiments, epigenetic modification reduces or halts the transcription of a target gene having a target sequence. In some embodiments, epigenetic modification reduces or halts the transcription of a copy of the target gene having a specific allele recognized by the epigenetic editor. In some embodiments, the epigenetic editor reduces or eliminates the expression level of the protein encoded by the target gene. In some embodiments, the epigenetic editor reduces or eliminates the expression level of the protein encoded by a copy of the target gene having a specific allele recognized by the epigenetic editor. The target PCSK9 gene can be epigenetically modified in vitro, exovivo, or in vivo.

[0197] The effector domains of the epigenetic editors described herein can alter (e.g., affix or remove) chemical modifications in nucleotides of a target gene or histones associated with the target gene. Chemical modifications may be altered in a single nucleotide or a single histone, or in 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1500, 2000, 2500, 3000, or more nucleotides.

[0198] In some embodiments, the effector domain of the epigenetic editor described herein can modify CpG dinucleotides within a target gene. In some embodiments, all CpG dinucleotides within 2000 bp, 1500 bp, 1000 bp, 500 bp, or 200 bp (e.g., within the modification sites described herein) adjacent to the target sequence are modified according to the type of modification described herein, compared to the gene in its original state or in an equivalent cell that has not come into contact with the epigenetic editor. In some embodiments, at least one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty, twenty-five, thirty, thirty, thirty, thirty, forty, forty, forty, fifteen In some embodiments, at least 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% of the CpG dinucleotide is modified compared to the original state of the gene or the gene in an equivalent cell that has not been in contact with the epigenetic editor. In some embodiments, a single CpG dinucleotide is modified compared to the original state of the gene or the gene in an equivalent cell that has not been in contact with the epigenetic editor.

[0199] The effector domains of the epigenetic editors described herein can alter the histone modification state of a histone associated with or bound to a target gene. For example, an effector domain can affix a modification to one or more lysine residues of the histone tail of a histone associated with a target gene. In some embodiments, the effector domain can result in deacetylation of one or more histone tails of a histone associated with a target gene, thereby reducing or silencing the expression of the target gene. In some embodiments, the histone modification state is a methylation state. For example, an effector domain can result in H3K9, H3K27, or H4K20 methylation (e.g., one or more of H3K9me2, H3K9me3, H3K27me2, H3K27me3, and H4K20me3 methylation) of one or more histone tails associated with a target gene, thereby reducing or silencing the expression of the target gene.

[0200] In some embodiments, all histone tails of histones bound to DNA nucleotides within 2000 bp, 1500 bp, 1000 bp, 500 bp, or 200 bp adjacent to the target sequence are modified according to the type of modification described herein, compared to the original state of the chromosome or to a chromosome in an equivalent cell that has not come into contact with the epigenetic editor. In some embodiments, at least one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty, twenty-five, thirty, thirty, thirty-five, forty, forty-five, fifty, fifty, fifty, sixty, sixty, sixty, seventy, seventy, eighty, eighty, ninety, ninety, fifty, fifty, sixty, sixty, seventy, seventy, eighty, eighty, ninety, ninety, fifty, fifty, one hundred In some embodiments, at least 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% of the histone tail of a bound histone is modified compared to the original state of the chromosome or the chromosome in an equivalent cell that has not come into contact with the epigenetic editor. For example, one single histone tail of a bound histone may be modified compared to the original state of the chromosome or the chromosome in an equivalent cell that has not come into contact with the epigenetic editor. As another example, one single bound histone octamer may be modified compared to the original state of the chromosome or the chromosome in an equivalent cell that has not come into contact with the epigenetic editor.

[0201] The chemical modifications conferred to target gene DNA nucleotides or histone residues may be located at or near the target sequence in the target gene. In some embodiments, the effector domain of the epigenetic editor described herein alters the chemical modification state of nucleotides or histone tails attached to nucleotides at the 5' or 3' end of the target sequence in the target gene by 100-200, 200-300, 300-400, 400-55, 500-600, 600-700, or 700-800 nucleotides. In some embodiments, the effector domain alters the chemical modification state of up to 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900, or 2000 nucleotides or histone tails bound to those nucleotides, adjacent to the target sequence. As used herein, “adjacent to” refers to a specific sequence, for example, nucleotide positions from the 5' end to the 5' end and from the 3' end to the 3' end of the target sequence.

[0202] In some embodiments, the effector domain mediates or induces a change in the chemical modification of a nucleotide distal to the target sequence or a histone tail bound to that nucleotide. Such modification may begin near the target sequence and subsequently extend to one or more nucleotides in the target gene distal to the target sequence. For example, the effector domain may initiate a change in the chemical modification state of one or more nucleotides or one or more histone residues bound to one or more nucleotides within 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, or 500 nucleotides adjacent to the target sequence, and the change in chemical modification state may extend from the target sequence within the target gene. The sequence may extend either upstream or downstream of the target sequence to one or more nucleotides, comprising at least 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900, 2000, 2500, 3000, or more nucleotides. In certain embodiments, chemical modification may begin at fewer than 2, 3, 5, 10, 20, 30, 40, 50, or 100 nucleotides in the target gene and extend to at least 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, or more nucleotides in the target gene. In some embodiments, the chemical modification extends to the entire nucleotide range of the target gene. Additional proteins or transcription factors, such as transcriptional repressors, methyltransferases, or transcriptional regulatory scaffold proteins, may be involved in the expansion of chemical modification. Alternatively, the epigenetic editor alone may be involved.

[0203] In some embodiments, the epigenetic editors described herein reduce the expression of a target gene by at least about 20%, at least about 30%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, at least about 90%, at least about 95%, at least about 99%, or more, when measured by transcription of the target gene in cells, tissues, or subjects, compared to control cells, control tissues, or control subjects (e.g., in the absence of the epigenetic editor). In some embodiments, the epigenetic editors described herein reduce the expression of a copy of the target gene by at least about 20%, at least about 30%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, at least about 90%, at least about 95%, at least about 99%, or more, when measured by transcription of a copy of the target gene in cells, tissues, or subjects, compared to control cells, control tissues, or control subjects. In certain embodiments, a copy of a target gene has a specific sequence or allele recognized by an epigenetic editor. In certain embodiments, the epigenetically modified copy encodes a functional protein, and therefore, the epigenetic editors disclosed herein can reduce or halt the expression and / or function of the protein.For example, the epigenetic editors described herein can reduce the expression and / or function of a protein encoded by a target gene in cells, tissues, or subjects to at least one-third, at least one-fifth, at least one-sixth, at least one-seventh, at least one-eighth, at least one-ninth, at least one-tenth, at least one-eleventh, at least one-twelfth, at least one-thirteenth, at least one-fourteenth, at least one-fifteenth, at least one-twentieth, at least one-twenty-fifth, at least one-thirtieth, at least one-thirtieth, at least one-fortieth, at least one-forty-fifth, at least one-fifty-fifty, at least one-sixtieth, at least one-seventieth, at least one-eightyth, at least one-ninetieth, or at least one-hundredth of a degree compared to control cells, control tissues, or control subjects.

[0204] The regulation of target gene expression can be assayed by determining any parameters that are indirectly or directly affected by the expression of the target gene. Such parameters include, for example, changes at the RNA or protein level, changes in protein activity, changes at the product level, changes in downstream gene expression, changes in transcription or activity of reporter genes such as luciferase, CAT, beta-galactosidase, or GFP, changes in signal transduction, changes in phosphorylation and dephosphorylation, changes in receptor-ligand interactions, and second messengers such as cGMP, cAMP, IP3, and Ca2 +Examples include changes in concentration, changes in cell growth, changes in angiogenesis, and / or changes in any functional effect of gene expression. Measurements may be performed in vitro, in vivo, and / or ex vivo, and may be performed by conventional methods, such as measurement of RNA or protein levels, measurement of RNA stability, and / or identification of downstream gene or reporter gene expression. Readouts may be performed by, for example, chemiluminescence, fluorescence, colorimetric reactions, antibody binding, inducible markers, ligand binding assays, changes in intracellular second messengers, such as cGMP and inositol triphosphate (IP3), changes in intracellular calcium levels, cytokine release, etc.

[0205] Methods for determining the expression level of a gene, such as an epigenetic editor target, may include determining the gene transcript level by reverse transcription PCR, quantitative RT-PCR, droplet digital PCR (ddPCR), Northern blotting, RNA sequencing, DNA sequencing (e.g., sequencing of complementary deoxyribonucleic acid (cDNA) obtained from RNA), next-generation sequencing, nanopore sequencing, pyrosequencing, or nanostring sequencing. The level of proteins expressed from a gene can be determined by, for example, Western blotting, enzyme-coupled immunosorbent assay, mass spectrometry, immunohistochemistry, or flow cytometry. The gene expression product level may be normalized against an internal standard, such as total messenger ribonucleic acid (mRNA), or against the expression level of a specific gene, such as a housekeeping gene.

[0206] In some embodiments, the effect of an epigenetic editor on regulating target gene expression can be tested using a reporter system. For example, an epigenetic editor may be designed to target a reporter protein, such as a reporter gene encoding a fluorescent protein. The expression of the reporter gene in such a model system can be monitored, for example, by flow cytometry, fluorescence-activated cell sorting (FACS), or fluorescence microscopy. In some embodiments, a cell population may be transfected with a vector containing the reporter gene. The vector may be constructed so that the reporter gene is expressed when the vector is transfected into cells. Suitable reporter genes include genes encoding fluorescent proteins, such as green, yellow, cherry, cyan, or orange fluorescent proteins. A cell population having a reporter system may be transfected with DNA, mRNA, or a vector encoding an epigenetic editor that targets the reporter gene.

[0207] VII. Pharmaceutical Compositions In one embodiment, the Disclosure provides a pharmaceutical composition comprising one or more epigenetic editors or components thereof (e.g., fusion proteins and / or guide polynucleotides) described herein as an active ingredient (or a single active ingredient), or a nucleic acid molecule(s) encoding such epigenetic editor or component(s). For example, the pharmaceutical composition may comprise a nucleic acid molecule(s) encoding a fusion protein(s) (and, where applicable, a guide polynucleotide) of an epigenetic editor described herein. In some embodiments, a separate pharmaceutical composition comprises a fusion protein(s) and a guide polynucleotide(s). The pharmaceutical composition may also comprise cells that have undergone epigenetic modifications(s) mediated or induced by the epigenetic editors provided herein.

[0208] In general, the epigenetic editors or their components(s) described herein, or the nucleic acid molecules(s) encoding such epigenetic editors or their components(s), are preferred to be administered as formulations together with one or more pharmaceutically acceptable excipients, for example, as described below.

[0209] The term “excipient” is used herein to describe any component other than the compound(s) of this disclosure. The choice of excipient(s) may largely depend on factors such as the specific mode of administration, the effect of the excipient on solubility and stability, and the nature of the dosage form. As used herein, “pharmaceutically acceptable excipients” include any physiologically compatible solvent, dispersion medium, coating agent, antimicrobial and antifungal agent, isotonic agent and absorption retarder. Some examples of pharmaceutically acceptable excipients are water, saline, phosphate-buffered saline, dextrose, glycerol, ethanol, and combinations thereof. Often, it is preferable to include an isotonic agent in the composition, such as sugar, polyhydric alcohol, such as mannitol, sorbitol, or sodium chloride. Examples of additional pharmaceutically acceptable substances are wetting agents or trace amounts of auxiliary substances, such as wetting or emulsifying agents, preservatives, or buffering agents, which enhance the shelf life or efficacy of the antibody.

[0210] Pharmaceutical compositions suitable for parenteral administration typically contain an active ingredient combined with a pharmaceutically acceptable carrier, such as sterile water or sterile isotonic saline. Such formulations may be prepared, packaged, or marketed in a form suitable for bolus or serial administration.

[0211] VIII. Delivery method In some embodiments, the epigenetic editor or its component(s) is introduced into target cells in the form of nucleic acid molecules(s) encoding the epigenetic editor or its component(s), and therefore the pharmaceutical compositions herein include nucleic acid molecules(s). Such nucleic acid molecules(s) may be, for example, DNA, RNA, or mRNA, and / or modified nucleic acid sequences(s) (e.g., having chemical modifications, a 5' cap, or one or more 3' modifications). In some embodiments, the nucleic acid molecules(s) may be delivered as naked DNA or RNA, for example, by transfection or electroporation, or conjugated to a molecule that promotes uptake by target cells (e.g., N-acetylgalactosamine). In some embodiments, the nucleic acid molecules(s) may be contained in a nucleic acid expression vector(s), which may include expression regulatory sequences, such as promoters, enhancers, transcription signal sequences, transcription termination sequences, introns, polyadenylation signals, Kozak consensus sequences, internal ribosome entry sites (IRES), etc. Such regulatory expression sequences are well known in the art. The vector may also include a sequence encoding a signal peptide (e.g., for nuclear, nucleolar, or mitochondrial localization) that is associated with (e.g., inserted into or fused to) the protein-coding sequence.

[0212] Examples of vectors include, but are not limited to, plasmid vectors, vaccinia virus, poliovirus, adenovirus, adeno-associated virus, SV40, herpes simplex virus, human immunodeficiency virus, viral vectors based on retroviruses (e.g., vectors derived from retroviruses such as mouse leukemia virus or splenic necrosis virus, Rous sarcoma virus, Harvey sarcoma virus, avian leukemia virus, lentivirus, human immunodeficiency virus, myeloproliferative sarcoma virus, and mammary tumor virus), and other recombinant vectors. In certain embodiments, the vector is a plasmid or a viral vector. Viral particles or virus-like particles (VLPs) can also be used to deliver nucleic acid molecules that encode the epigenetic editors or their components(s) described herein. For example, “empty” viral particles can be assembled to contain any suitable cargo. Viral vectors and viral particles can also be manipulated to incorporate targeted ligands to alter their target tissue specificity.

[0213] In certain embodiments, the epigenetic editor or component(s) described herein is encoded by nucleic acid sequences(s) present in one or more viral vectors or suitable capsid proteins of any viral vector. Examples of viral vectors include adeno-associated virus vectors (e.g., derived from AAV3, AAV3b, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh8, AAV10, and / or their variants), retroviral vectors (e.g., Maloney's mouse leukemia virus, MML-V), adenovirus vectors (e.g., AD100), lentiviral vectors (e.g., vectors based on HIV and FIV), and herpesvirus vectors (e.g., HSV-2).

[0214] In some embodiments, delivery involves an adeno-associated virus (AAV) vector. AAV vector delivery may be particularly useful when the DNA-binding domain of the epigenetic editor fusion protein is a zinc finger array. While we do not wish to be constrained by any particular theory, the smaller size of a zinc finger array compared to a large DNA-binding domain, such as a Cas protein domain, may allow for the convenient loading of such fusion proteins into viral vectors, such as AAV vectors.

[0215] Any AAV serotype, such as human AAV serotypes, can be used in the AAV vectors described herein, including, but are not limited to, AAV serotype 1 (AAV1), AAV serotype 2 (AAV2), AAV serotype 3 (AAV3), AAV serotype 4 (AAV4), AAV serotype 5 (AAV5), AAV serotype 6 (AAV6), AAV serotype 7 (AAV7), AAV serotype 8 (AAV8), AAV serotype 9 (AAV9), AAV serotype 10 (AAV10), and AAV serotype 11 (AAV11), as well as their variants. In some embodiments, the AAV variants have at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% amino acid sequence identity with wild-type AAV. In certain embodiments, an AAV variant may be manipulated so that its capsid protein has reduced immunogenicity or enhanced transduction ability in humans. In some cases, a chimeric variant is generated by shuffling and reassembling one or more regions of at least two different AAV serotype viruses. For example, a chimeric AAV may contain terminal inverted repeat sequences (ITRs) that belong to a serotype that is heterologous to the capsid serotype. The resulting chimeric AAV may have different antigenic reactivity or recognition compared to its parent serotype. In some embodiments, a chimeric variant of AAV contains amino acid sequences derived from two, three, four, five, or more different AAV serotypes.

[0216] Nonviral systems are also intended for delivery as described herein. Nonviral systems include, but are not limited to, nucleic acid transfection methods including electroporation, sonoporation, calcium phosphate transfection, microinjection, DNA gene gun, transfection via heat shock, transfection mediated by small DNA, lipofection, transfection mediated by cationic agents, and transfection in liposomes, immunoliposomes, exosomes, or cationic facial amphiphiles (CFAs). In certain embodiments, one or more mRNAs encoding an epigenetic editor fusion protein as described herein may be co-electroporated with one or more guide polynucleotides (e.g., gRNAs) as described herein.

[0217] Any type of cell can be targeted for delivery of the epigenetic editor or its component(s) described herein. For example, the cell may be a eukaryotic cell or a prokaryotic cell. In some embodiments, the cell is a mammalian (e.g., human) cell. Examples of human cells include hepatocytes, cholangiocarcinoma (cholangiocarcinoma), stellate cells, Kupffer cells, and hepatic sinusoidal endothelial cells.

[0218] In some embodiments, the epigenetic editors or their components(s) described herein are delivered to host cells for transient expression, for example, via a transient expression vector. Transient expression of the epigenetic editors or their components(s) may result in long-term or permanent epigenetic modification of the target gene. For example, the epigenetic modification may be stable for at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 weeks or longer after the introduction of the epigenetic editor into host cells, or for 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 months or longer. The epigenetic modification may be maintained after one or more mitotic and / or meiotic events in the host cell. In certain embodiments, the epigenetic modification is maintained across generations in offspring arising from or derived from the host cell.

[0219] IX. Therapeutic use of epigenetic editors This disclosure also provides a method for treating or preventing a condition in a subject, comprising the step of administering to the subject an epigenetic editor or pharmaceutical composition described herein. The epigenetic editor brings about an epigenetic modification of a target polynucleotide sequence within a target gene associated with a disease, condition, or disorder in the subject, thereby regulating the expression of the target gene and treating or preventing the disease, condition, or disorder. In some embodiments, the epigenetic editor reduces the expression of the target gene to a degree sufficient to achieve a desired effect, e.g., a therapeutically relevant effect, e.g., prevention or treatment of a disease, condition, or disorder.

[0220] In some embodiments, a subject is administered a system for modulating (e.g., suppressing) the expression of PCSK9, wherein the system comprises (1) a fusion protein(s) of an epigenetic editor described herein and, if associated, a guide polynucleotide(s), or (2) a nucleic acid molecule encoding the fusion protein(s) and, if associated, a guide polynucleotide(s).

[0221] "To treat," "to treat," and "treatment" refer to a method of alleviating or suppressing at least one of a biological disorder and / or its associated symptoms. As used herein, "to alleviate" a disease, disorder, or condition means reducing the severity and / or frequency of the symptoms of the disease, disorder, or condition. Furthermore, references to "treatment" herein include curative, temporary palliative, and preventive treatments. In some embodiments, alleviating symptoms compared to an equivalent untreated control may include a reduction of at least 3%, 5%, 10%, 20%, 40%, 50%, 60%, 80%, 90%, 95%, 98%, 99%, 99.5%, 99.9%, or 100% of symptoms, as measured by any standard technique.

[0222] In some embodiments, the subject may be a mammal, such as a human. In some embodiments, the subject may be selected from non-human primates, such as chimpanzees, crab-eating macaques, macaques, and other ape and monkey species.

[0223] In some embodiments, the human patient has a condition selected from hypercholesterolemia (e.g., familial hypercholesterolemia, e.g., heterozygous familial hypercholesterolemia (HeFH) or homozygous familial hypercholesterolemia (HoFH), or established atherosclerotic cardiovascular disease (ASCVD)) or renal insufficiency (RI).

[0224] In some embodiments, the patient to be treated with the epigenetic editor of the present disclosure has received prior treatment for the condition to be treated (e.g., hypercholesterolemia (e.g., HeFH, HoFH, HF, or established ASCVD) or RI). In other embodiments, the patient has not received such prior treatment. In some embodiments, the patient has failed prior treatment for the condition (e.g., prior hypercholesterolemia treatment).

[0225] The epigenetic editor of the present disclosure can be administered to a patient having a condition described herein in a therapeutically effective amount. As used herein, "therapeutically effective amount" refers to the amount of the therapeutic agent being administered that will, to some extent, alleviate one or more of the symptoms of the disorder being treated and / or result in the clinical endpoint(s) desired by the medical practitioner. The effective amount of a therapy can be measured by its ability to stabilize disease progression and / or alleviate symptoms in a patient, preferably to reverse disease progression. The ability of the epigenetic editor of the present disclosure to reduce or silence PCSK9 expression can be evaluated, for example, by in vitro assays described herein, as well as in suitable animal models that predict efficacy in humans. A suitable dosing regimen is selected to provide an optimal therapeutic response in each specific situation, and can be administered, for example, as a single bolus or continuous infusion, with dose adjustment as appropriate based on the urgency of each case.

[0226] The epigenetic editor of the present disclosure can be administered without additional therapeutic treatment, i.e., as a stand-alone therapy (monotherapy). Alternatively, treatment with the epigenetic editor of the present disclosure may include at least one additional therapeutic treatment (combination therapy). In some embodiments, the additional therapeutic agent is any one known in the art for treating hypercholesterolemia or RI. Therapeutic agents include, but are not limited to, statins, fibrate drugs, HMG-CoA reductase inhibitors, niacin, bile acid modulators or adsorbents, cholesterol absorption inhibitors or modulators, CETP inhibitors, MTTP inhibitors, and PPAR agonists.

[0227] The epigenetic editor of the present disclosure or its components (or a nucleic acid molecule encoding the epigenetic editor or its components) can be administered by any method acceptable in the art, for example, subcutaneously, intradermally, intratumorally, intranodally, intramuscularly, intravenously, intralymphatically, or intraperitoneally. In certain embodiments, the pharmaceutical composition of the present disclosure is administered intravenously to a subject.

[0228] X. Definitions As used herein, the term “nucleic acid” means any oligonucleotide or polynucleotide comprising a nucleotide (e.g., deoxyribonucleotide or ribonucleotide) in either single-stranded or double-stranded form, including DNA and RNA. A “nucleotide” comprises the sugar deoxyribose (DNA) or ribose (RNA), a base, and a phosphate group linked together via the phosphate group. The “base” includes, but is not limited to, naturally occurring compounds such as adenine, thymine, guanine, cytosine, uracil, inosine, and purines and pyrimidines, as well as modified versions of purines and pyrimidines with novel reactive groups, such as amines, alcohols, thiols, carboxylic acids, alkyl halides, etc. Nucleic acids may include known nucleotide analogs and / or modified skeletal residues or linkages, which may be synthetic, naturally occurring, or not naturally occurring. Such nucleotide analogs, modified residues, and modified ligatures are well known in the art and may result in enhanced cellular uptake of nucleic acid molecules, reduced immunogenicity, and / or increased stability in the presence of nucleases.

[0229] As used herein, “isolated” or “purified” nucleic acid molecules are nucleic acid molecules that exist apart from their natural environment. For example, an “isolated” or “purified” nucleic acid molecule is (1) separated from the nucleic acid of its origin, genomic DNA or cellular RNA, and / or (2) not found in nature. In some embodiments, an “isolated” or “purified” nucleic acid molecule is a recombinant nucleic acid molecule.

[0230] In addition to the specific proteins and nucleic acid molecules referred to herein, it will be understood that this disclosure also intends to utilize their variants, derivatives, homologs, and fragments. A variant of any given sequence may have a specific sequence of residues (whether amino acid residues or nucleic acid residues) modified in such a manner that the polypeptide or polynucleotide substantially retains at least one of its endogenous functions. Variant sequences can be obtained by the addition, deletion, substitution, modification, replacement, and / or variation of at least one residue present in naturally occurring sequences (in some embodiments, one or fewer, two or fewer, three or fewer, four or fewer, five or fewer, six or fewer, seven or fewer, eight or fewer, nine or fewer, ten or fewer, fifteen or fewer, or twenty or fewer residues). With respect to certain proteins described herein (e.g., the KRAB, dCas9, DNMT3A, and DNMT3L proteins described herein), this disclosure also intends to include any variants or homologs that retain at least one of the naturally occurring forms of the protein or its endogenous function (e.g., at least 50%, 60%, 70%, 80%, 90%, 85%, 96%, 97%, 98%, or 99% of the function compared to the specific protein described herein).

[0231] Several exemplary fusion proteins, which are encompassed by this disclosure, are provided herein. Those skilled in the art will understand that these exemplary proteins are non-limiting examples and that additional proteins are included within the scope of this disclosure. For example, if an exemplary fusion protein is provided that includes a particular domain, e.g., a mammalian DNMT3A, DNMT3L, and / or KRAB domain, e.g., a human or mouse DNMT3A, DNMT3L, and / or KRAB domain, then those skilled in the art will be able to see that, in some embodiments, fusion proteins having the same configuration but in which one or more mammalian domains are replaced with homology domains derived from another mammal, e.g., one or more mouse domains are replaced with one or more human domains, are also encompassed by this disclosure. For example, if an exemplary fusion protein containing a mouse DNMT3L domain is provided, then fusion proteins having the same configuration but in which mouse DNMT3L is replaced with a human DNMT3L domain are also encompassed.

[0232] When used herein, homologs of any polypeptide or nucleic acid sequence contemplated herein include sequences having certain homology to wild-type amino acids and nucleic acid sequences. Homologous sequences may include sequences that are at least 50%, 55%, 65%, 75%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the sequence of interest, such as amino acid sequences. In the context of amino acid or nucleotide sequences, the term "~ percent identical" refers to the percentage of residues that are the same in the two sequences when aligned with respect to the greatest match. In some embodiments, the length of the reference sequence aligned for comparison purposes is at least 30% of the reference sequence (e.g., at least 40%, 50%, 60%, 70%, 80%, or 90%, or 100%). Sequence identity can be measured using sequence analysis software (e.g., Genetics Computer Group's Sequence Analysis Software Package, University of Wisconsin Biotechnology Center, 1710 University Avenue, Madison, Wis. 53705, BLAST, BESTFIT, GAP, or PILEUP / PRETTYBOX program). Such software associates identical or similar sequences by assigning a degree of homology to various substitutions, deletions, and / or other modifications. In an exemplary approach to determining the degree of identity, the BLAST program may be used, where probability scores e-3 to e-100 indicate closely related sequences.

[0233] The percentage of identity between two nucleotide or polypeptide sequences is determined, for example, using BLAST® with default parameters (available on the US National Library of Medicine's National Center for Biotechnology Information website). In some embodiments, the length of the reference sequence aligned for comparison is at least 30% of the reference sequence (e.g., at least 40, 50, 60, 70, 80, or 90%).

[0234] It will be understood that the numbering of specific positions or residues in a polypeptide sequence depends on the specific protein and numbering scheme used. Numbering may differ, for example, between the precursor of a mature protein and the mature protein itself, and interspecies sequence differences can affect numbering. Those skilled in the art will be able to identify any homologous protein and the respective residues in the nucleic acids encoding them by methods well known in the art, such as sequence alignment and homology residue determination.

[0235] The terms “modulate” or “alter” refer to a change in the quantity, degree, or scope of a function. For example, the epigenetic editors described herein can modulate the activity of a promoter sequence by binding to a motif within the promoter, thereby inducing, enhancing, or repressing the transcription of a gene operably ligated to the promoter sequence. As another example, the epigenetic editors described herein can block RNA polymerase from transcribing a gene or inhibit the translation of an mRNA transcript. The terms “inhibit,” “suppress,” “suppress,” and “silence,” when used in reference to the epigenetic editors or components thereof described herein, refer to reducing or preventing the activity (e.g., transcription) of a nucleic acid sequence (e.g., a target gene) or protein compared to the activity of the nucleic acid sequence or protein in the absence of the epigenetic editor or component thereof. This term may include partially or completely blocking activity, or preventing or delaying activity. The inhibited activity may be, for example, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% lower than that of the control, or it may be, for example, at least 1 / 1.5, 1 / 2, 1 / 3, 1 / 4, 1 / 5, or 1 / 10 of that of the control.

[0236] The terms “approximately” or “about” mean within the acceptable margin of error for a particular value as determined by those skilled in the art, which will depend in part on how the value is measured or determined, for example, on the limitations of the measuring system. For example, “approximately” may mean within a standard deviation of 1 or greater than 1, according to convention for a given value. Where a particular value is described in this application and claims, unless otherwise indicated, the term “approximately” should be assumed to mean within the acceptable margin of error for that particular value.

[0237] It is understood that the ranges provided herein are all abbreviated forms of the values ​​within that range. For example, the range 1 to 50 is understood to include any digits, combinations of digits, or subranges from the group consisting of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50, as well as all intervening decimal values ​​between the aforementioned integers, such as 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, and 1.9. With respect to subranges, "nested subranges" extending from any endpoint of the range are specifically intended. For example, nested subranges of an exemplary range of 1 to 50 may include 1 to 10, 1 to 20, 1 to 30, and 1 to 40 in one direction, or 50 to 40, 50 to 30, 50 to 20, and 50 to 10 in the other direction.

[0238] Unless otherwise defined herein, scientific and technical terms used in connection with this disclosure shall have the meanings generally understood by those skilled in the art. Exemplary methods and materials are described below, but similar or equivalent methods and materials may also be used in the practice or testing of this disclosure. In case of any conflict, including definitions, this specification shall prevail. Furthermore, unless otherwise required by context, singular terms shall include plural forms, and plural terms shall include singular forms. Throughout this specification and its embodiments, the words “have” and “comprise,” or variations such as “has,” “having,” “comprises,” or “comprising,” shall mean the inclusion of an integer or group of integers indicated, but not the exclusion of any other integer or group of integers. Unless otherwise indicated, descriptions of lists of elements herein include any of the elements individually or in any combination. The descriptions of embodiments in this specification include such embodiments as single embodiments or in combination with any other embodiments herein. All publications, patents, patent applications, and other references referenced herein are incorporated in their entirety by reference. To the extent that any reference incorporated by reference conflicts with the disclosure contained herein, this specification is intended to take precedence and / or be superior to any such conflicting material. Several documents are cited herein, but this citation does not constitute an understanding that any of these documents form part of the common general knowledge in the art.

[0239] According to this disclosure, backreferences in dependent claims mean abbreviations for the direct and clear disclosure of each individual claim combination indicated by the backreference. Furthermore, headings in this specification are created for ease of organization and are not intended in any way to limit the scope of methods and compositions described in the claims.

[0240] Some protein sequences provided herein, for example, some fusion protein sequences, include peptide tags, such as His6 tags or DYKDDDDK (SEQ ID NO: 1528) tags, which are useful for the detection and / or purification of tagged proteins but do not affect the function of the protein. These peptide tags and additional suitable peptide tags are well known to those skilled in the art. It will be apparent to those skilled in the art that the disclosed tags can be substituted with other suitable peptide tags, and that fusion proteins with the same or highly similar sequences but without such peptide tags, for example, those created with cleaved or no peptide tags, are equally suitable for carrying out the embodiments of the present disclosure.

[0241] To better understand this disclosure, the following examples are provided. These examples are illustrative only and should not be construed as limiting the scope of this disclosure.

[0242] [Examples] [Example 1] Design and synthesis of fusion proteins We designed and constructed a fusion protein ("CRISPR-off") containing dCas9, DNMT3A, DNMT3L, and KOX1 KRAB. From the N-terminus to the C-terminus, this protein has the domain DNMT3A-linker-DNMT3L-XTEN80-NLS-dSpCas9-NLS-XTEN16-KOX1 KRAB (SEQ ID NOs. 658 and 1495). The CRISPR-off plasmid construct is described in Nunez (Nunez et al., Cell (2021) 184(9):2503-19) and was ordered from Twist Biosciences.

[0243] We also constructed a ZF fusion protein ("ZF-off") containing DNMT3A, 3L, and KOX1 KRAB. This construct has the general structure DNMT3A-linker-DNMT3L-XTEN80-NLS-ZFP domain-NLS-XTEN16-KOX1Krab (SEQ ID NOs. 659 and 1496).

[0244] [Example 2] Selection of target PCSK9 sequences for gRNA epigenetic silencing gRNAs targeting ±1kb from the PCSK9 TSS were designed by computer computation using the Benchling gRNA platform (Benchling (2021), available from benchling.com) for human (GRCh38), mouse (mm10), and cynomolgus monkey (Macaca fascicularis) (5.0) PCSK9. gRNAs containing poly-TTTT sequences were discarded first. The inventors performed gRNA off-target analysis using CasOFFinder (Bae et al., Bioinformatics (2014) 30(10):1473-5). gRNAs were discarded if they matched multiple locations across the respective genomes constructed for each independent species.

[0245] Cross-reactivity sequencing analysis was performed on human PCSK9 gRNA to annotate sequence mismatches with cynomolgus monkey or mouse gRNA sequences. Specifically, gRNA sequence alignment was performed to identify the degree of DNA similarity at each nucleotide, including annotation of guides containing up to 0, 1, or 2 nucleotide mismatches. The final set of 226 gRNA sequences was selected for primary PCSK9 screening in HeLa cells.

[0246] [Example 3] Selection of ZF target sites and design of ZF proteins for epigenetic silencing Using a library of two-finger ZFPs (2F units), each recognizing a 6bp DNA site, a larger six-finger ZFP array targeting an 18bp DNA binding site was designed. The source of the 2F units was a set of three-finger zinc finger proteins selected to bind to specific target sites using a bacterial-2-hybrid (B2H) selection system (Hurt et al., PNAS (2003) 100:12271-6, Maeder et al., Mol Cell (2008) 31(2):294-301). By generating all possible triplet combinations of the 6bp binding sites represented in the library and allowing either 0 or 1bp between the 6bp target sites, a list of targetable DNA sites was created. To identify zinc finger target sites within PCSK9, ±1kb sequences from TSS (human (GRCh38)) were matched against this list. For each identified ZF target site, multiple ZF proteins could be designed. Six recognition helices were designed to generate the complete protein by selecting a two-finger unit and considering several factors, such as the known binding preference of zinc finger proteins, the frequency with which amino acids at positions 1, 2, 3, and 6 are selected to bind to the desired target base in the B2H selection system, the avoidance of amino acids at positions 1, 2, 3, and 6 selected to bind to multiple different bases in B2H, and maintaining context dependence by matching adjacent bases where possible. The complete ZF sequence was derived from the naturally occurring Zif268 protein, and the selected recognition helices were maintained in the sequence context selected in B2H (finger 1-2 or finger 2-3 of Zif268). The two-finger unit was linked using the linker TGSQKP (SEQ ID NO: 651) when the 6 bp binding site was contiguous, and using the linker TGGGSQKP (SEQ ID NO: 652) when the 6 bp binding site was 1 bp separated. A final set of 209 ZFPs, targeted to 49 different binding sites, was selected for the primary PCSK9 screening in HeLa cells.

[0247] Figure 1 shows the overlap of gRNAs and zinc finger proteins mapped to the PCSK9 target region.

[0248] [Example 4] Screening of guide RNAs in HeLa cells Primary screening of gRNAs targeting PCSK9 was performed in HeLa cells. The gRNA sequences were ordered from Twist Biosciences as DNA fragments having a u6 promoter sequence in front of the gRNA coding sequence.

[0249] HeLa cells were transfected with gRNA and CRISPR-off in DNA form. Six 96-well plates (Sigma-Aldrich catalog no. M2936) were seeded with 12,000 HeLa cells per well (ATCC catalog no. CCL-2) in standard culture medium containing 10% fetal bovine serum (Thermo Fisher catalog no. A4766) volume / volume, 1× GlutaMAX™ (Thermo Fisher catalog no. 35050061), and 1× penicillin-streptomycin (Thermo Fisher catalog no. 15140122) supplemented with DMEM (Thermo Fisher catalog no. 11-965-092). After plating, the cells were grown for 24 hours in an incubator at 37°C and 5% CO2. 25 ng of each gRNA fragment and 50 ng of the CRISPR-off (SEQ ID NO: 658) plasmid were resuspended in DPBS buffer (Thermo Fisher catalog number 14190144) to a concentration of 7.5 ng / μL. In addition, 10 ng of the EF1a:puromycin-resistant plasmid was also added to the transfection mix to bring the total payload to 85 ng of DNA. Transfection mixtures were prepared by adding the resuspended DNA components to Mirus® TransIT®-LT1 transfection reagent (Mirus catalog number MIR2300) according to the manufacturer's instructions. 10 μL of each transfection mixture was added in double layers to a total of six screening plates. The positive control used was CRISPRi (dCas9-KRAB) with two gRNA targeting sites adjacent to the TSS. The two CRISPRi-positive control gRNAs used were gRNA004 and gRNA005, which are annotated in the gRNA sequence table. These control conditions are referred to as "CRISPRi-1" and "CRISPRi-2," respectively, in the table of primary screening data.Negative controls included CRISPR-off without gRNA, CRISPR-off using gRNA targeting a non-PCSK9 locus (CD151), and an empty vector (pUC19, NEB catalog number N3041S).

[0250] Puromycin resistance selection was performed 24 hours after transfection. The cell medium was completely aspirated, all wells were washed three times with DPBS buffer (Thermo Fisher catalog number 14190144), and 200 μL of 1 μg / μL puromycin was added to all wells of the screening plate.

[0251] Cells were passaged 48 hours after transfection. The cell medium was completely aspirated, and all wells were washed three times with DPBS buffer (Thermo Fisher catalog no. 14190144). Cells were enzymatically lifted by adding 25 μL of trypsin-EDTA (0.25%) (Thermo Fisher catalog no. 25200056) in a 37°C incubator for 5 minutes. The trypsin-treated cells were resuspended in fresh standard culture medium at a 1:8 ratio and replated at a 1:4 ratio 72 hours after the medium change.

[0252] To measure the levels of secreted PCSK9 protein, culture medium was collected 24 hours after a medium change, and cell plates were assayed for relative cell count using the Promega Cell Titer Glo® protocol (catalog no. G7570) as recommended by the manufacturer. PCSK9 protein levels were evaluated using the LEGEND MAX® Human PCSK9 ELISA Kit (catalog no. 443107) from BioLegend. Collected medium was plated, and all subsequent steps were performed strictly according to the manufacturer's recommendations. Final plate readings were taken at 450 nm using Perkin Elmer®, VICTOR®, and Nivo® F instruments. Functions were fitted to standard curves and unknowns were interpolated using GraphPad Prism software. To correct for variability in cell count per well, the PCSK9 ELISA results were normalized by the Cell Titer Glo® assay (Promega catalog no. G7571) results to compensate for any variability in cell count per well.

[0253] Over 200 gRNAs were tested, and 40 of them were identified as top sequences (Figure 2, top sequences are indicated by dark circles). The sequences and efficacy of the tested gRNAs are shown in Table 7 (SEQ: Sequence ID). Relative PCSK9 secretion ("PCSK9% relative to control") represents the mean PCSK9 protein level of treated samples, expressed as a percentage of the mean across all non-targeted gRNA (CD151) negative control conditions. Strong PCSK9 silencing (30-40% relative to negative control levels) was observed in cells treated with several candidate gRNAs. The top 40 gRNAs with the best PCSK9 protein knockdown were selected to be ordered as sgRNAs for further follow-up studies. Table 7. Target sequences of the tested gRNAs

[0254] [Table 7] TIFF2026510339000033.tif255165TIFF2026510339000034.tif255165TIFF2026510339 000035.tif255165TIFF2026510339000036.tif255165TIFF2026510339000037.tif62165

[0255] The most functional gRNA was found to be aligned near the PCSK9 gene transcription start site (Figure 3).

[0256] Following this primary screening, a secondary screening was performed using the top 40 gRNAs in RNA form. The top 40 guides were chemically synthesized and co-transfected with in vitro transcribed mRNA encoding CRISPR-off, CRISPRi, or wild-type Cas9 constructs. Secreted PCSK9 levels were measured 7 and 28 days after transfection.

[0257] To generate in vitro transcribed CRISPR-off, CRISPRi, and wild-type Cas9 effector mRNAs, plasmid constructs encoding these proteins were linearized using MfeI restriction enzyme from NEB® (catalog no. R3589S). Using 1 μg of the linearized template, the in vitro transcription reaction was set up using CellScript's T7 mScript® standard mRNA production system (catalog no. C-MSC100625) according to the manufacturer's instructions. The resulting RNA had a cap 1 structure at the 5' end and was polyadenylated at 3'. The transcribed RNA was purified using Qiagen's RNeasy® Mini kit (catalog no. 74104).

[0258] Terminally modified sgRNAs, purified using standard desalting methods, were obtained from Integrated DNA Technologies. Three nucleotides at the 5' end and three nucleotides at the 3' end of each guide were 2'-O-methyl modified. Three nucleoside junctions at the 3' end and three nucleoside junctions at the 5' end were phosphorothioate nucleoside junctions (Table 8, SEQ: Sequence ID). In this table, mX (i.e., mA, mC, mG, or mU) represents a 2'-O-methyl modified ribonucleoside, rX (i.e., rA, rC, rG, or rU) represents a native ribonucleoside, and * indicates a phosphorothioate junction. All nucleoside junctions that are not phosphorothioate junctions are phosphate junctions. Table 8. Target sequences of the top 40 gRNAs obtained from secondary HeLa cell screening.

[0259] [Table 8] TIFF2026510339000039.tif230165TIFF2026510339000040.tif230165TIFF2026510339000041.tif230165 TIFF2026510339000042.tif230165TIFF2026510339000043.tif230165TIFF2026510339000044.tif159165

[0260] HeLa cells were reverse-transfected with 25 ng of effector and 12.5 ng of sgRNA in 96-well plates using the TransIT®-X2 transfection reagent from Mirus (catalog number MIR6003). The conditioned medium was collected weekly for up to 4 weeks, and secreted PCSK9 levels were measured using the LEGEND MAX® Human PCSK9 ELISA kit from BioLegend (catalog number 443107). ELISA data were normalized to cell number using the CellTiter-Glo® kit from Promega (catalog number G7571). PCSK9 silencing was transient with CRISPRi (dCas9-KRAB), returning to baseline by day 28, but several sgRNAs co-transfected with the CRISPR-off (DNMT3A-3L-dCas9-KRAB) construct showed potent and persistent silencing (Figure 4A). Of the 40 guides tested using CRISPR-off, 16 showed superior silencing efficiency compared to wild-type Cas9 (Table 9). Table 9. Relative PCSK9 expression in modified gRNA and CRISPR-off treated HeLa cells.

[0261] [Table 9] TIFF2026510339000046.tif47158

[0262] At the 2-week mark, RNA was extracted using the Quick-RNA 96 kit (catalog number R1053) from Zymo Research. qPCR was performed using the qScript XLT One-Step RT-qPCR ToughMix (catalog number 95134-500) and TaqMan assay (PCSK9: Hs00545399_m1, PPIA: Hs99999904_m1) from Quantabio. PCSK9 levels were normalized with PPIA. Relative quantification was performed using the delta-delta Ct method.

[0263] Suppression of PCSK9 secretion was found to correlate with mRNA silencing on day 14 (Figure 4B).

[0264] modRNA004 and modRNA111 were tested in HeLa cells for suppressing PCSK9 secretion over 60 days (Figure 5). Wild-type Cas9 was co-transfected with modRNA180 as a positive control. Cells were treated with 25 ng of effector and 12.5 ng of gRNA. modRNA004 and modRNA111 were shown to mediate sustained silencing of PCSK9, which was equivalent to that achieved by wild-type Cas9 via gene editing in HeLa cells.

[0265] Furthermore, the results suggest that epigenetic silencing is maintained in simvastatin-treated HeLa cells. Statin treatment is known to increase PCSK9 secretion via a transcriptional mechanism. In epigenetically silenced HeLa cells, statin treatment was shown not to increase PCSK9 secretion (Figure 6).

[0266] [Example 5] Guide RNA assay in Huh7 hepatoma cell line The Huh7 hepatoma cell line is suitable for high-throughput screening and transfection. The top 13 guides obtained from HeLa screening, possessing either complete homology or a single mismatch with the cynomolgus monkey PCSK9 gene, were tested in the Huh7 hepatoma cell line. Epigenetic silencing in the CRISPR-off construct was shown to be stable for 7 days (Figure 7).

[0267] Next, the specificity of fusion proteins 11 and 13 using PCSK9-targeted gRNA 041 (see Table 12) was measured in Huh7 cells. Changes in gene expression were evaluated by RNA sequencing of treated and untreated cells. The values ​​were visualized in a volcano plot (Figure 25). No significant off-target effects were detected.

[0268] [Example 6] Guide RNA assay in primary human and cynomolgus monkey hepatocytes The efficacy of gRNA in primary hepatocytes will be tested using primary human and cynomolgus monkey HepatoPac® cultures from BioIVT. HepatoPac® cultures will be maintained according to the manufacturer's recommendations. Briefly, HepatoPac® maintenance medium will be thawed and prepared within 30 minutes of cell arrival. During medium changes, cells will be acclimatized for 2 days in an incubator at 37°C and 10% CO2. On day 2 after receipt, CRISPR-off+ sgRNA, GFP-mRNA, and wild-type CRISPR-Cas9 will be administered to primary cultures at various concentrations using standard procedures. The medium will be changed every other day for a period of up to 4 weeks to assess the persistence and / or inheritability of silencing. PCSK9 silencing will be assessed every 7 days by measuring the level of PCSK9 secreted into the medium by ELISA. PCSK9 concentration is controlled against total hepatocytes using human albumin ELISA (Thermo Fisher®). Data are then presented as total PCSK9 secretion, as a percentage relative to a GFP-mRNA-negative control. Specificity can be assessed by isolating primary human and cynomolgus monkey hepatocytes from a mouse fibroblast feeder layer using a magnetic bead-based antibody approach (Miltenyi Biotec). After isolation of primary hepatocytes from the feeder layer, the cells are processed for RNA-seq evaluation and whole-genome bisulfite sequencing.

[0269] The top 13 gRNAs in RNA form will be selected for testing in primary human hepatocytes (PHH). The top gRNAs will be selected based on (i) PCSK9 silencing efficiency and persistence in HeLa cells, and (ii) complete alignment with the human PCSK9 gene and whether they have up to one mismatch with the non-human primate PCSK9 gene. GRNA combinations will also be tested to determine their efficacy and persistence. Negative controls will be CRISPR-off only, gRNA fragment only (modRNA003), and CRISPRi only. The positive control will be CRISPRi co-transfected with modRNA004 (Table 10, SEQ: Sequence ID, NHP: Non-human primate). All tested gRNAs are expected to bind to both human and non-human primate PCSK9. Table 10. gRNAs screened in primary human hepatocytes

[0270] [Table 10] TIFF2026510339000048.tif68165

[0271] Strong PCSK9 silencing is observed. For some gRNAs, a reduction of more than 70% of secreted PCSK9 is observed on day 7 (depending on transfection efficiency).

[0272] [Example 7] ZF assay in HeLa cells A total of 209 zinc finger proteins (the composition of which is shown in SEQ ID NO: 659) were designed from a ZF library targeting 49 PCSK9 target sites (selected from GRCh38 chromosome 1, 55038548–55040548). The target sites did not have any other strict matches in the human genome (GRCh38).

[0273] HeLa cells were transfected with the ZF-off construct in DNA form. Six 96-well plates (Sigma-Aldrich catalog no. M2936) were seeded with 12,000 HeLa cells (ATCC catalog no. CCL-2) per well in standard culture medium containing 10% fetal bovine serum (Thermo Fisher catalog no. A4766) volume / volume, 1× GlutaMAX™ (Thermo Fisher catalog no. 35050061), and 1× penicillin-streptomycin (Thermo Fisher catalog no. 15140122) supplemented with DMEM (Thermo Fisher catalog no. 11-965-092). After plating, the cells were grown in a 37°C incubator at 5% CO2 for 24 hours. 10 ng of ZF-off plasmid was resuspended in DPBS buffer (Thermo Fisher catalog number 14190144) to a concentration of 7.5 ng / μL. In addition, 10 ng of EF1a:puromycin-resistant plasmid and 65 ng of empty vector (pUC19) were also added to the transfection mix to a total payload of 85 ng of DNA. The resuspended DNA was added to serum-free OPTI-MEM medium (Thermo Fisher® catalog number 31985062), and the transfection mixture was prepared by adding Mirus® TransIT®-LT1 transfection reagent (MIR2300) according to the manufacturer's instructions. 10 μL of the transfection mixture was added in double layers to a total of six screening plates. The positive control was CRISPR-off (SEQ ID NO: 658) using high-performance gRNA (gRNA009). Negative controls included ZF-off and empty vectors (pUC19, NEB catalog number N3041S) using non-PCSK9 gene coordinate (CLTA).

[0274] ZF screening yielded hits with activity equivalent to CRISPR (Figure 8). Candidates with high silencing efficiency were advanced to follow-up experiments. Figure 9 shows the ZF screening results by distance to TSS. A total of 209 ZFs were screened, and their PCSK9 knockdown activity compared to the negative control is shown in Table 11 below. Table 11. ZF-off construct activity

[0275] [Table 11] TIFF2026510339000050.tif161159

[0276] Several ZF-off constructs were shown to be more effective than CRISPR-off in combination with wild-type Cas9 and gRNA003 for silencing PCSK9. The target sites of the ZFP domains and ZF sequences (F1-F6) in these ZF-off constructs are shown in Table 1.

[0277] [Example 8] Complete specificity screening of constructs in primary human hepatocytes The specificity of CRISPR-off and ZF-off constructs for silencing PCSK9 will be tested in primary human hepatocytes. Readouts for evaluating specificity will include RNA-seq, methylation arrays, and whole-genome bisulfite sequencing assays. Changes in genome-wide expression and methylation after epigenetic editing will be profiled compared to negative controls.

[0278] The specificity of five gRNAs was tested in PXB cells. Novel human hepatocytes were isolated from PXB mouse model mice (Figure 26A). Long-term stability and functionality of the hepatocytes, as well as potent PCSK9 secretion, were confirmed. Cells were treated with the epigenetic repressor PLA2628 (fusion protein 12 in Example 12) or a control. PCSK9 secretion was measured and plotted as a percentage of PCSK9 produced and secreted by the negative control (PLA2628 using PCSK9 non-targeting gRNA ("off-target")) (the time course is shown in Figure 26B, and the level at day 14 is shown in Figure 27). Construct specificity was tested by RNA-seq for all gRNAs 14 days after delivery. Volcano plots for the two exemplary RNAs evaluated (RNA041 and RNA049, sequences shown in Table 12) are shown in Figure 26C. No significant off-target effects were observed for any of the five tested gRNAs.

[0279] [Example 9] CpG methylation patterns The CpG methylation patterns in human hepatocytes (e.g., primary cells or cell lines) treated with CRISPR-off or ZF-off will be investigated. A hybrid capture assay will be performed on bisulfite-treated DNA to examine the methylation patterns at CpG sites induced by CRISPR-off or ZF-off in a 1kb region surrounding the PCSK9 TSS.

[0280] In mice treated with CRISPR-off (gRNA 041 and 049) or ZF-off (ZFP152) constructs, a whole-liver hybrid capture assay was performed 90 days after partial hepatectomy and compared with untreated or vehicle-only control mice, as described in more detail below. Methylation was examined across a genomic region of approximately 5kb, including the PCSK9 promoter region. Low levels of CpG methylation were observed in untreated and vehicle-only control mice, while significant levels of CpG methylation were observed in CRISPR-off and ZF-off treated mice (data not shown).

[0281] [Example 10] Stable PCSK9 silencing by epigenetic editing in mice with wild-type PCSK9 The ability of CRISPR-off and ZF-off constructs to mediate epigenetic silencing of endogenous PCSK9 in vivo will be tested. Constructs will be delivered using a single intravenous dose of mRNA (and gRNA for CRISPR-off silencing). Silencing will be tested in wild-type mice over a period of 2–6 months. Readouts will be serum PCSK9 levels and serum cholesterol levels. Subsets of each cohort will be selected for liver hematoxylin and eosin (H&E) staining and RNA-seq analysis. For several constructs, potent, stable, and inheritable PCSK9 silencing will be observed.

[0282] [Example 11] Stable PCSK9 silencing by epigenetic editing in mice expressing transgenic human PCSK9 Three different mouse strains expressing transgenic human PCSK9 are available: hPCSK9-Tg(mPCSK9+ / -) heterozygous mice, hPCSK9-Tg(mPCSK9+ / +) homozygous mice, and hPCSK9-Tg(mPCSK9- / -) mice. The available hPCSK9-Tg(mPCSK9- / -) mouse strain is C57BL / 6J-Pcsk9- / -Tg(RP11-55M23-AbsI), which expresses human PCSK9 under the control of its own promoter (Figure 10). See, for example, Weider et al., J Biol Chem (2016) 291(32):16659-71.

[0283] CRISPR-off and ZF-off constructs were tested in hPCSK9-Tg (mPCSK9- / -) mice expressing hPCSK9. The constructs used were CRISPR / wtCas9 (SEQ ID NO: 2) using gRNA g079, CRISPRi (NLS-NLS-dCas9-NLS-KOX1KRAB-NLS-NLS) using gRNA g041, CRISPR-OFF PLA2628 (fusion protein 12 provided in Example 12 of this specification (SEQ ID NO: 1519)) using gRNA g056, CRISPR-OFF PLA2628 using gRNA g041 and g049, CRISPR-OFF PLA1489 (fusion protein 11 provided in Example 12 of this specification (SEQ ID NO: 1517)) using gRNA g041 and g049, and the ZF-OFF ZFP152ADD shown below.

[0284] ZFP152ADD (NLS-NLS-Dnmt3A-ADD-hDnmt3L-ZF-ZNF627KRAB-NLS-NLS): TIFF2026510339000051.tif94162

[0285] Constructs were delivered by a single intravenous administration of an epigenetic silencer (Figure 20B), and each mouse was given a 3 mg / kg dose of the formulation. The selected guide RNAs used are shown in Table 12. The efficacy of PCSK9 silencing was measured by serum ELISA for human PCSK9 at 42 days (Figure 21A) and 84 days (Figure 21B) after intravenous administration. Mice treated with constructs that did not show sustained PCSK9 silencing were sacrificed at week 6. Sustained silencing was observed with PLA2628 used with gRNA g041 and g049, PLA1489 used with gRNA g041 and g049, and ZFP152ADD. Table 12. Guide RNAs tested in transgenic mice

[0286] [Table 12]

[0287] In further experiments, g041 and g49 were synthesized with the following chemical modification patterns (where "m" refers to a ribonucleotide modified with 2'-OMe, "r" refers to an unmodified ribonucleotide, and "*" refers to a phosphorothioate linkage).

[0288] TIFF2026510339000053.tif30165

[0289] TIFF2026510339000054.tif30165

[0290] Modified guide RNAs were tested in hPCSK9-Tg (mPCSK9- / -) mice expressing hPCSK9. The guides were administered individually (mg041 or mg049) or together (mg041 and mg049) along with mRNA constructs encoding the following epigenetic repressor proteins.

[0291] TIFF2026510339000055.tif21165TIFF2026510339000056.tif141165

[0292] Each mouse was administered a single dose of epigenetic repressor mRNA and gRNA in a 1:1 ratio at 0.375 mg / kg to deliver construct and guide RNAs. The efficacy of PCSK9 silencing was measured at 7 and 14 days post-administration by plasma ELISA against human PCSK9. Each identical sequence of terminally modified guide (having three 5' terminal and three 3' terminal nucleotides modified with 2'-OMe and phosphorothioate linkages) was used as a positive control. Sustained silencing of PCSK9 was observed for all guides, and comparable silencing was observed between the terminally modified control guide RNA and the tested modified guide RNA at both time points. Similar silencing levels were observed at both time points tested.

[0293] [Example 12] Fusion protein with variant NLS configuration We developed several improved fusion protein constructs using variant nuclear localization sequence (NLS) configurations to achieve significantly higher episilencing activity.

[0294] Several constructs with variant NLS domain configurations (Figures 11A and 11B) were constructed and tested at the PCSK9 locus in HeLa cells (Figures 12A-12B). The constructs were further tested at the PCSK9 locus in Hepa1-6 (Figure 13) and HuH7 (Figures 14A-14C and 15). The amino acid and DNA sequences of exemplary fusion protein constructs are shown below.

[0295] [Table 12-2] TIFF2026510339000058.tif242170TIFF2026510339000059.tif242170TIFF2026510339000060.tif242170TIFF2026510339000061.tif242170TIFF2026510339000062.tif242170TIFF2026510339000063.tif242170TIFF2026510339000064.tif242170TIFF2026510339000065.tif242170TIFF2026510339000066.tif242170TIFF2026510339000067.tif242170TIFF2026510339000068.tif242170TIFF2026510339000069.tif242170TIFF2026510339000070.tif242170TIFF2026510339000071.tif241170TIFF2026510339000072.tif241170TIFF2026510339000073.tif241170TIFF2026510339000074.tif241170TIFF2026510339000075.tif241170TIFF2026510339000076.tif241170TIFF2026510339000077.tif241170TIFF2026510339000078.tif241170TIFF2026510339000079.tif241170TIFF2026510339000080.tif241170TIFF2026510339000081.tif242170TIFF2026510339000082.tif242170TIFF2026510339000083.tif242170TIFF2026510339000084.tif242170TIFF2026510339000085.tif242170TIFF2026510339000086.tif242170TIFF2026510339000087.tif242170TIFF2026510339000088.tif242170TIFF2026510339000089.tif242170TIFF2026510339000090.tif242170TIFF2026510339000091.tif242170TIFF2026510339000092.tif242170TIFF2026510339000093.tif242170TIFF2026510339000094.tif32170.

[0296] The sequence from Figure 14A can be found below:

[0297] [Table 12-3]

[0298] The sequence from Figure 15 can be found below:

[0299] [Table 12-4] TIFF2026510339000097.tif239170TIFF2026510339000098.tif239170TIFF2026510339000099.tif124170

[0300] Cell culture and transfection HeLa (ATCC-CRM-CCL-2), Hepa1-6 (PCSK9-IRES-TdTomato), Huh7 (Sekisui XenoTech, LLC), and HEK293T Griptite (CLTA-GFP) cells were cultured in DMEM with 10% FBS. All experiments with HeLa and Huh7 cells were performed using chemically synthesized guide RNA and in vitro transcribed effector constructs. HeLa cells were reverse-transfected using TransIT-X2 transfection reagent from Mirus (catalog no. MIR6003). Huh7 cells were reverse-transfected using MessengerMAX reagent from Invitrogen (catalog no. LMRNA003). Secreted PCSK9 levels were measured at indicated time points using the LEGEND MAX® Human PCSK9 ELISA kit from Biolegend (catalog no. 443107). All ELISA data were normalized to cell number using the CellTiter-Glo kit from Promega (catalog number G7571).

[0301] HEK293T Griptite cells, in which GFP was knocked into the CLTA locus as an in-frame CLTA fusion, were co-transfected with plasmids encoding the effector construct and human CLTA guide RNA using the TransIT-X2 transfection reagent from Mirus (catalog number MIR6003). GFP expression was measured by FACS as a surrogate for CLTA expression.

[0302] Hepa1-6 cells were co-transfected with an effector construct and a plasmid encoding mouse PCSK9 guide RNA using the SF Cell Line 96-well nucleofector kit (catalog number V4SC-2096, program code: CM-138) in an Amaxa 4D nucleofector device from Lonza. At the indicated time points, cells were analyzed by FACS with TdTomato expression as a surrogate for PCSK9 levels.

[0303] In vitro transcription of effector constructs and synthetic gRNAs. Using a 1 μg linearization effector template, an in vitro transcription reaction was set up using the T7 mScript® standard mRNA production system from CellScript (catalog no. C-MSC100625) according to the manufacturer's instructions to obtain RNA having a cap 1 structure at the 5' end and polyadenylated at 3'. Terminally modified sgRNAs having three 2'O-methyl modified nucleotides with phosphorothioate linkages at both the 5' and 3' ends were obtained from Integrated DNA Technologies.

[0304] Methylation profiling Genomic DNA was extracted from each well of a 96-well culture plate using the DNAdvance DNA Extraction from Tissue kit (Beckman Coulter). After quantification of genomic DNA using the High Sensitivity DNA 1X kit (Quant-IT), each genomic DNA sample was bisulfite converted using the EZ-96 DNA Methylation-Gold MagPrep kit (Zymo Research) according to the manufacturer's instructions. For hybridization capture experiments, DNA libraries were prepared using the xGen® Methyl-Seq DNA Library Prep kit (IDT), and hybrid capture was performed using the xGen® DNA Library Hybridization Capture Kit (IDT). For amplicon sequencing experiments, DNA libraries were prepared using the xGen® Methyl-Seq DNA Library Prep kit (IDT), and hybrid capture was performed using the xGen® DNA Library Hybridization Capture Kit (IDT). Using bisulfite-converted DNA obtained from each sample, PCRs corresponding to each of the two VIM amplicons were seeded using the Platinum Taq kit (Invitrogen). The pooled products were cleaned using the AMPure XP kit (Beckman Coulter), and fragment size evaluation was performed using D1000 screen tape on a Tapestation 4200 (Agilent) before sequencing with a commercially available service (Azenta).

[0305] [Example 13] Bacterial DNA methyltransferase In this experiment, a panel of bacterial proteins was screened for DNA methyltransferase activity in mammalian cells. These bacterial DNA methyltransferases (Table 13) were tested for their epigenetic silencing activity by fusing them to a dCas9 domain at the N-terminus using the experimental procedure of Example 1. These constructs were then transfected into GFP-expressing reporter cell lines under the control of the mammalian promoter CTLA4. Table 13. Bacterial DNA methyltransferases

[0306] [Table 13]

[0307] M.SssI DNA methyltransferase efficiently methylates DNA in mammalian cells (Figure 16), and the silencing was stable for up to 30 days. The sequence from Figure 16 can be found below:

[0308] [Table 13-2]

[0309] Furthermore, the methylation profiles of these cells were analyzed on day 29, and methylation of 20% of the target genes was confirmed (Figures 17-18).

[0310] Three other orthologous DNA methyltransferases predicted to be closely related to M. SssI were identified and tested for epigenetic silencing activity using the experimental procedure of Example 1 (Table 14). Table 14. Bacterial DNA methyltransferases

[0311] [Table 14]

[0312] The DNA methyltransferases in Table 14 are predicted to have similar or improved functions to those of M. SssI. The sequences will be tested under CRISPR-off conditions instead of mouse DNMT3A / DNMT3L, and their functions will be compared to those of M. SssI DNA methyltransferase in silencing the PCSK9 locus in the HeLa TdTomato system to identify novel features and improved functions.

[0313] [Example 14] Alternative KRAB domain In this example, the fusion protein was constructed using an alternative KRAB domain (Table 15), which showed improved activity compared to CRISPR-off when tested using the experimental procedure of Example 1 (Figures 19A-19D). Table 15. Alternative KRAB domains

[0314] [Table 15] TIFF2026510339000104.tif191170

[0315] [Example 15] ZIM 3 Fusion Construct This study aims to generate a novel fusion of ZIM3 and KOX1KRAB. Both ZIM3 and KOX1KRAB are KRAB family proteins with broad homology. The study designs sequences representing a halfway point between ZIM3 and KOX1KRAB. These KOX1KRAB and ZIM3 constructs encode small regions of KOX1KRAB and ZIM3 concentrated around the zinc finger domains of the proteins. While the regions used by KOX1KRAB and ZIM3 are very similar within the first approximately 75 bp of their sequences, ZIM3 also has a small alpha-helix region at its C-terminus that is not present in KOX1KRAB. The KOX1KRAB-FL sequence contains the KOX1KRAB sequence corresponding to this extra piece, while the ZIM3 truncated form is obtained by removing this extra piece from the ZIM3 sequence. The ZIM3 / KOX1KRAB chimera is a fusion of the N-terminal and C-terminal pieces of the two proteins. All ZIM3-like KOX1KRAB variants were assembled by first assembling the 100 closest homologs ("families") of each gene using BLAST of ZIM3 or KOX1KRAB proteins from non-human species; secondly, identifying three members of the KOX1KRAB family most similar to ZIM3 and three members of the ZIM3 family most similar to KOX1KRAB; and thirdly, logically modifying the KOX1KRAB-FL sequence to resemble each of the three sets (Table 16). Table 16. ZIM-KOX1KRAB Chimeric Protein

[0316] [Table 16]

[0317] [Example 16] Dose-response experiment of PCSK9 silencing In mice expressing human PCSK9, various doses of CRISPR-off and ZF-off constructs were tested. Constructs that showed sustained PCSK9 silencing at 3 mg / kg (see Example 11) were selected for further testing. These included PLA2628 using gRNA g041 and g049, PLA1489 using gRNA g041 and g049, and ZFP152ADD, as disclosed in Example 11. Constructs were delivered as described in Example 11. Each target construct was tested at 0.2 mg / kg, 0.375 mg / kg, 0.75 mg / kg, and 3 mg / kg. Baseline hPCSK9 levels were determined two weeks prior to construct administration (Figure 22). The effectiveness of PCSK9 silencing was measured over 28 days. The results for PLA1489 and ZFP152ADD are shown in Figures 23A and 23B, respectively. The figures show human PCSK9 serum levels after intravenous administration, as measured by ELISA. Sustained silencing of hPCSK9 to levels below approximately 10% of baseline was observed for each construct at several test doses.

[0318] [Example 17] PCSK9 silencing persists even after partial hepatectomy. Partial hepatectomy is an established method for inducing liver regeneration in mice and has been used to test the persistence of genetic and epigenetic effects in mice. Mice were treated with a PCSK9-targeted epigenetic editor using a single intravenous dose. Eleven mice were each given either (1) a placebo dose (vehicle only - negative control), (2) 1.5 mg / kg of the epigenetic editor, or (3) 3 mg / kg of the epigenetic editor. The mice were then divided into cohorts A and B. After 3 months, mice in cohort B had 70 percent of their livers surgically removed, while mice in cohort A did not undergo surgery. After 2 months, the livers of the mice in cohort B had regenerated. Serum levels of PCSK9 were measured periodically from post-injection in both cohorts (Figure 24). In both Cohort A and Cohort B, mouse PCSK9 silencing persisted throughout 140 days post-injection.

[0319] array The sequence numbers (SEQs) of the nucleotide (nt) and amino acid (aa) sequences described in this disclosure are listed below.

[0320] [Table 17] TIFF2026510339000107.tif251164TIFF2026510339000108.tif251164TIFF2026510339000109.tif251164TIFF2026510339000110.tif251164TI FF2026510339000111.tif251164TIFF2026510339000112.tif251164TIFF2026510339000113.tif251164TIFF2026510339000114.tif251164TIFF2 026510339000115.tif250164TIFF2026510339000116.tif250164TIFF2026510339000117.tif251164TIFF2026510339000118.tif251164TIFF202 6510339000119.tif251164TIFF2026510339000120.tif253163TIFF2026510339000121.tif253163TIFF2026510339000122.tif253163TIFF202651 0339000123.tif253163TIFF2026510339000124.tif253163TIFF2026510339000125.tif253163TIFF2026510339000126.tif253163TIFF2026510339000127.tif250164TIFF2026510339000128.tif250164TIFF2026510339000129.tif252164TIFF2026510339000130.tif254164TIFF20265103390 00131.tif254164TIFF2026510339000132.tif254164TIFF2026510339000133.tif254164TIFF2026510339000134.tif254164TIFF2026510339000135.tif254164TIFF2026510339000136.tif254164TIFF2026510339000137.tif254164TIFF2026510339000138.tif254164TIFF2026510339000139.tif254164TIFF2026510339000140.tif254164TIFF2026510339000141.tif254164TIFF2026510339000142.tif254164TIFF2026510339000143.tif254164TIFF2026510339000144.tif254164TIFF202651033900 0145.tif254164TIFF2026510339000146.tif254164TIFF2026510339000147.tif254164TIFF2026510339000148.tif254164TIFF2026510339000149.tif254164TIFF2026510339000150.tif254164TIFF2026510 339000151.tif254164TIFF2026510339000152.tif254164TIFF2026510339000153.tif254164TIFF2026510339000154.tif254164TIFF2026510339000155.tif254164TIFF2026510339000156.tif254164TIFF20 26510339000157.tif254164TIFF2026510339000158.tif249164TIFF2026510339000159.tif253164TIFF2026510339000160.tif249164TIFF2026510339000161.tif252163TIFF2026510339000162.tif197164.

Claims

1. A system for suppressing the transcription of the human PCSK9 gene in human cells, and in some cases in human liver cells, a) A domain of DNA methyltransferase (DNMT) and / or a domain that recruits DNMT, wherein the DNMT domain and / or recruiter domain may include a DNMT3A domain and / or a DNMT3L domain, and the recruited DNMT may be DNMT3A, and Transcriptional repressor domain It collectively includes, Each domain is linked to a DNA-binding domain that binds to a target sequence in the human PCSK9 gene, and the target sequence includes one or more fusion proteins containing the sequences of SEQ ID NO: 687, SEQ ID NO: 1039, SEQ ID NO: 1044, or SEQ ID NO: 1046, or b) One or more nucleic acid molecules encoding one or more fusion proteins A system that includes this.

2. The system according to claim 1, wherein the DNA-binding domain comprises a dead CRISPR-Cas (dCas) domain, and the system comprises a guide RNA that targets the fusion protein to one or more sequences in the PCSK9 gene selected from SEQ ID NOs. 1039, 1044, and 1046.

3. The system according to claim 2, wherein the system comprises (i) one or more guide RNAs including any one of sequence numbers 1491 to 1493, or (ii) one or more nucleic acid molecules encoding one or more guide RNAs of (i).

4. Guide RNA, sequence mA * mC * mU * rGrCrCrUrGrGrCrUrCrArCrUrCrCrUrCrCrGrUrUrUrUrArGrArGmCmUmAmGmAmAmAmUmAmGmCrArArGrUrUrArArArArArU rArArGrGrCrUrArGrUrCrCrGrUrUrArUrCrAmAmCmUmUmGmAmAmAmAmAmGmUmGrGmCmAmCmCmGmAmGmUmCmGmGmUmGmCmU * mU * mU * The system according to claim 3, comprising mU (mg041, sequence number 1528).

5. The guide RNA has the sequence mA * mU * mC * rGrUrCrCrGrArUrGrGrGrGrCrUrCrUrGrGrGrUrUrUrUrArGrArGmCmUmAmGmAmAmAmUmAmGmCrArArGrUrUrArArArArUrArArGrGrCrUrArGrUrCrCrGrUrUrArUrCrAmAmCmUmUmGmAmAmAmAmAmGmUmGrGmCmAmCmCmGmAmGmUmCmGmGmUmGmCmU * mU * mU * The system according to claim 3, comprising mU (mg049, SEQ ID NO: 1529).

6. The system according to any one of claims 2 to 5, wherein the dCas domain comprises a sequence having at least 90% identity with the dCas9 sequence, and optionally sequence number 12 or 13.

7. The system according to claim 1, wherein the DNA-binding domain includes a ZFP domain that binds to a nucleotide target sequence containing the sequence of sequence number 687.

8. The system according to claim 7, wherein the ZFP domain comprises the F1-F6 amino acid sequence of ZF034 shown in Table 1.

9. The system according to any one of claims 1 to 8, wherein the DNMT3A domain comprises a sequence having at least 90% identity with sequence number 574 or 575.

10. The system according to any one of claims 1 to 9, wherein the DNMT3L domain comprises a sequence having at least 90% identity with a sequence selected from sequence numbers 578 to 581.

11. The system according to any one of claims 1 to 9, wherein the DNMT3L domain comprises a sequence having at least 90% identity with a sequence selected from sequence numbers 582 to 603.

12. The system according to any one of claims 1 to 8, wherein the DNMT domain comprises a sequence having at least 90% identity with a sequence selected from sequence numbers 601 to 603.

13. The system according to any one of claims 1 to 12, wherein the transcriptional repressor domain comprises a sequence having at least 90% identity with a sequence selected from sequence numbers 33 to 570.

14. The system according to any one of claims 1 to 12, wherein the transcriptional repressor domain comprises a KRAB domain derived from KOX1, ZIM3, ZFP28, or ZN627.

15. The system according to claim 14, wherein the KRAB domain comprises a sequence having at least 90% identity with a sequence selected from sequence numbers 89, 116, 245, and 255.

16. The system according to any one of claims 1 to 12, wherein the transcriptional repressor domain comprises a fusion of the N-terminal and C-terminal regions of ZIM3 and KOX1 KRAB, and optionally comprises the amino acid sequence of SEQ ID NO: 571 or 572.

17. The system according to any one of claims 1 to 12, wherein the transcriptional repressor domain is derived from KAP1, MECP2, HP1a / CBX5, HP1b, CBX8, CDYL2, TOX, TOX3, TOX4, EED, EZH2, RBBP4, RCOR1, or SCML2.

18. The system a) comprising a DNMT3A domain, a DNMT3L domain, a transcriptional repressor domain, and a DNA-binding domain, In some cases, one or both of the DNMT3A and DNMT3L domains are human. A fusion protein in which the DNA-binding domain is a dead CRISPR-Cas domain or a ZFP domain, b) Nucleic acid molecules encoding a fusion protein A system according to any one of claims 1 to 17, including the system described in any one of claims 1 to 17.

19. The system according to claim 18, wherein the fusion protein comprises, from the N-terminus to the C-terminus, a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a DNA-binding domain, a third peptide linker, and a transcriptional repressor domain.

20. The system according to claim 19, wherein the fusion protein comprises, from N-terminus to C-terminus, a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a first nuclear localization signal (NLS), a DNA binding domain, a second NLS, a third peptide linker, and a transcriptional repressor domain.

21. The system according to claim 18, wherein the fusion protein comprises, from N-terminus to C-terminus, a first nuclear localization signal (NLS), a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a DNA-binding domain, a third peptide linker, a transcriptional repressor domain, and a second NLS.

22. The system according to claim 18, wherein the fusion protein comprises, from N-terminus to C-terminus, first and second nuclear localization signals (NLS), a DNMT3A domain, a first peptide linker, a DNMT3L domain, a second peptide linker, a DNA-binding domain, a third peptide linker, a transcriptional repressor domain, and third and fourth NLS.

23. The system according to any one of claims 18 to 22, wherein the transcriptional repressor domain is a KRAB domain, and optionally a human KOX1, ZFP28, ZN627, or ZIM3 KRAB domain.

24. The system according to any one of claims 19 to 23, wherein one or both of the second and third peptide linkers are XTEN linkers, optionally selected from XTEN80 and XTEN16, and optionally the second peptide linker is XTEN80 and the third peptide linker is XTEN16.

25. The system according to claim 18, wherein the fusion protein comprises, from N-terminus to C-terminus, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a first NLS, a dSpCas9 domain, a second NLS, an XTEN16 peptide linker, and a human KOX1 KRAB domain.

26. The system according to any one of claims 1 to 6 or 9 to 15, wherein the fusion protein comprises a sequence that is at least 90% identical to sequence number 1519.

27. The system according to claim 1 or any one of claims 7 to 15, wherein the fusion protein comprises the sequence of SEQ ID NO: 688 or a sequence that is at least 90% identical thereto, and the F1 to F6 amino acid sequence of ZF034 shown in Table 1.

28. The system according to claim 1 or any one of claims 7 to 15, wherein the fusion protein comprises the sequence of SEQ ID NO: 689 or a sequence that is at least 90% identical thereto, and the F1 to F6 amino acid sequence of ZF034 shown in Table 1.

29. The system according to claim 1 or any one of claims 7 to 15, wherein the fusion protein comprises a sequence of sequence number 690 or a sequence that is at least 90% identical thereto, and the F1 to F6 amino acid sequence of ZF034 shown in Table 1.

30. The system according to claim 1 or any one of claims 7 to 15, wherein the fusion protein comprises a sequence of sequence number 1527 or a sequence that is at least 90% identical thereto, and the F1 to F6 amino acid sequence of ZF034 shown in Table 1.

31. The system according to claim 18, wherein the fusion protein comprises, from N-terminus to C-terminus, a human DNMT3A domain, a first peptide linker, a human DNMT3L domain, an XTEN80 peptide linker, a first NLS, a ZFP domain, a second NLS, an XTEN16 linker, and a human KOX1 KRAB domain.

32. The system according to claim 31, wherein the fusion protein comprises the sequence of SEQ ID NO: 659 or a sequence that is at least 90% identical thereto, or the sequence of SEQ ID NO: 1496 or a sequence that is at least 90% identical thereto, and optionally the ZFP comprises the F1-F6 amino acid sequence of ZF034 shown in Table 1.

33. The system according to claim 18, wherein the fusion protein comprises, from N-terminus to C-terminus, first and second NLS, human DNMT3A domain, first peptide linker, human DNMT3L domain, XTEN80 peptide linker, ZFP domain, XTEN16 peptide linker, human KOX1 KRAB domain, and third and fourth NLS, and optionally the fusion protein comprises a sequence that is at least 90% identical to SEQ ID NO: 1514 or thereto, or a sequence that is at least 90% identical to SEQ ID NO: 1525 or thereto.

34. The system according to claim 18, wherein the fusion protein comprises, from N-terminus to C-terminus, first and second NLS, human DNMT3A domain, first peptide linker, human DNMT3L domain, XTEN80 peptide linker, dSpCas9 domain, XTEN16 peptide linker, human ZIM3 KRAB domain, and third and fourth NLS.

35. The system according to claim 34, wherein the fusion protein comprises a sequence that is at least 90% identical to or from sequence number 698.

36. A human cell or cell offspring comprising the system described in any one of claims 1 to 35, wherein the cell is optionally a hepatocyte.

37. A pharmaceutical composition comprising the system according to any one of claims 1 to 35, and a pharmaceutically acceptable excipient.

38. A method for treating a patient in need of treatment, comprising the step of administering to the patient, optionally intravenously, a system according to any one of claims 1 to 35 or a pharmaceutical composition according to claim 37.

39. The patient, Having heart disease, Having elevated low-density lipoprotein cholesterol (LDL-C) or hypercholesterolemia, At risk of developing myocardial infarction, stroke, or unstable angina, and / or Primary hyperlipidemia, and in some cases heterozygous familial hypercholesterolemia (HeFH) or homozygous familial hypercholesterolemia (HoFH), The method according to claim 38.

40. A system according to any one of claims 1 to 35 or a pharmaceutical composition according to claim 37 for use in the method according to claim 38 or 39, in part, in treating a patient in need of treatment.

41. Use of the system according to any one of claims 1 to 35 in the manufacture of a pharmaceutical product for treating a patient in need of treatment, optionally in the manner of claim 38 or 39.

Images