WO2025037261A1 - Methods of preventing or treating liver disease - Google Patents

Abstract

Methods of preventing or treating liver disease in a subject characterised as having a PNPLA3- 148M protein variant are provided. The methods include administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant. The disclosure also provides methods of editing a target ribonucleic acid (RNA) molecule encoding a PNPLA3-148M protein variant and methods of restoring the function of the PNPLA3 protein in a subject characterised as having the PNPLA3-148M protein variant.

Description

METHODS OF PREVENTING OR TREATING LIVER DISEASE

CROSS REFERENCE TO RELATED APPLICATION

This application claims priority to U.S. Provisional Patent Application No. 63/519,925, filed on 16 August 2023, titled “Methods of Preventing or Treating Liver Disease”, which is herein incorporated by reference in its entirety.

FIELD OF THE INVENTION

The present invention relates to methods of preventing or treating liver disease in a subject characterised as having a PNPLA3-148M protein variant. The methods comprise administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the subsequently edited polynucleotide encodes a PNPLA3-148V protein variant. In further aspects, the invention provides methods of editing a target ribonucleic acid (RNA) molecule encoding a PNPLA3-148M protein variant and methods of restoring the function of the PNPLA3 protein in a subject characterised as having the PNPLA3-148M protein variant.

BACKGROUND OF THE INVENTION

Liver diseases are a significant cause of mortality worldwide. Non-alcoholic fatty liver disease (NAFLD) is a common liver disorder that is prevalent in 20-30% of the global population (Salari et al. 2021 ; Unalp-Arida and Ruhl 2020). NAFLD is caused by irregular accumulation of fat in liver cells and may develop into non-alcoholic steatohepatitis (NASH). Characteristics of NASH include hepatocyte damage, hepatocyte swelling, inflammatory infiltrate and the early development of perisinusoidal fibrosis (Xiang et al. 2021). NAFLD and NASH can also lead to cirrhosis, liver cancer, cardiovascular comorbidity and liver transplantation. Cirrhosis and liver cancer account for 3.5% of the global mortality rate (Xiang et al. 2021). Note that NAFLD has been renamed to “metabolic dysfunction associated steatotic liver disease” or “MAFLD.” Likewise, NASH has been renamed to “metabolic dysfunction associated steatohepatitis” or “MASH.” NAFLD and MAFLD are used interchangably herein. NASH and MASH are used interchangeably herein. Diagnosis of NAFLD and NASH requires the exclusion of excessive alcohol consumption and viral infection (Unalp-Arida and Ruhl 2020). The patatin-like phospholipase domain-containing 3 (PNPLA3) protein contains a patatin-like phospholipase domain and is highly expressed in liver and with some detectable levels in adipose tissues. In the liver, PNPLA3 protein is predominantly expressed in hepatocytes and stellate cells, where it resides on the surface of lipid droplets. The PNPLA3 protein has lipase activity and can hydrolyse triglyceride, acting on monounsaturated and polyunsaturated fatty acids, possibly promoting lipid remodeling involving transfer of polyunsaturated fatty acids from triglyceride to phospholipids (Xiang et al, 2021).

A known variant, PNPLA3-148M, results from the single nucleotide polymorphism (SNP) rs738409 (C>G), involving a missense mutation of cytosine to guanine. PNPLA3-148M has been shown to be highly associated with the occurrence and severity of NAFLD/NASH (Dai et aL, 2019; Xiang et al. 2021), resulting in greater inflammatory infiltration and liver damage in subjects compared to wild-type PNPLA3 subjects (Salari et al. 2021). The SNP rs738409 (C>G, PNPLA3-148M) is also reported to be associated with alcohol-related liver diseases (ArLD) (Xiang et al., 2021 ; Kolla et al., 2018; Stickel et aL, 2015). Alcohol consumption contributes to alcohol-related liver disease (ArLD). ArLD can develop into alcoholic steatohepatitis (ASH), a disease accompanied by liver inflammation. Long-term ASH eventually leads to fibrosis, cirrhosis, and liver transplantation, and can even develop into hepatocellular carcinoma (HOC) (Xiang et al. 2021). rs738409 is also significantly associated with HBV/HCV (hepatitis B virus/hepatitis C virus) progression, hepatocellular carcinoma, and liver transplant (Hsueh et al. 2022; Fan et al. 2016). Analysis of UK Biobank data indicated that for various ancestral groups, between 21 -74% of the population carries at least one copy of the risk allele rs738409 (C>G) and 24% of certain demographic groups are homozygous for the risk allele. It is estimated that more than 2 million NASH patients worldwide are homozygous for the risk allele rs738490 (C>G). It is estimated that more than 2 million NASH patients worldwide are homozygous for the risk allele rs738490 (C>G).

The disease variant PNPLA3-148M has multiple changes in function compared to the wild-type PNPLA3 protein. In hepatocytes, PNPLA3-148M has reduced lipase activity (loss of function activity) causing increased lipid droplet size (Huang et al., 2011 ; Dong, 2019). In addition, PNPLA3-148M has greater stability than wild type protein and accumulates to a greater extent on the surface of lipid droplets. This enhanced level of PNPLA3-148M suppresses adipose triglyceride lipase (ATGL, aka PNPLA2) activity by sequestering the shared co-activating factor CGI-58 (Wang et al., 2019; Dong 2019). These changes are believed to contribute to increased lipid droplet size and reduced lipolysis in hepatocytes, resulting in certain gain of function phenotypes. Thus, PNPLA3-148M is both a loss of function mutation (i.e., reduced enzymatic, specifically lipase, activity) and a gain of function mutation (i.e., increased accumulation of PNPLA3-148M protein on lipid droplets and increased lipid accumulation). In hepatic stellate cells, PNPLA3 has retinyl-palmitate lipase activity, and the PNPLA3-148M loss of function variant leads to retinol retention in the cells with increased secretion of pro-fibrogenic proteins and proinflammatory cytokines (Pingitore and Romeo, 2019; Pirazzi et al., 2014; Bruschi et al., 2017).

At present, there are no effective, approved therapies available for the treatment of NAFLD or NASH. There have been attempts to target the PNPLA3-148M disease variant as a therapeutic strategy. However, current strategies have sought to degrade or remove the disease variant, for instance by silencing the mutant allele (Linden et al. 2019; US National Library NIH clinical trial identifiers NCT05809934 and NCT05648214). This type of approach has various drawbacks. In particular, silencing the mutant allele fails to restore the normal function of the PNPLA3 wild-type protein.

Gene therapy techniques, in particular DNA editing techniques such as CRISPR-Cas, have been contemplated to edit the risk allele rs738409 (C>G). However, such techniques are associated with various risks. DNA editing permanently alters genomic information and there exist safety concerns of oncogenic off-target effects. Thus, such editing techniques may be less acceptable for broad therapeutic application.

SUMMARY OF THE INVENTION

There remains a need for safe, effective therapies for liver disease including new effective therapies for preventing and/or treating NASH (or MASH). The present invention addresses this need by providing methods based on targeting the PNPLA3-148M disease variant. More specifically, the present invention is based on the use of ADAR-mediated RNA editing to convert the codon “AUG” encoding the methionine at position 148 of PNPLA3-148M to codon “IUG”, which is read by ribosome during translation as GUG to encode a valine residue at position 148 (see Figure 1). Thus, this invention has the potential of producing a protein containing an amino acid replacement at position 148 with valine, an amino acid containing a structurally similar, hydrophobic alkyl sidechain isostere of the wild type isoleucine residue. Hepatic cells (HuH-7, primary hepatocytes, or iPS-cell derived hepatocytes) overexpressing the PNPLA3-148V variant were found to have decreased lipid content as compared with cells overexpressing the disease variant PNPLA3-148M (FIGs. 7A-7C; FIGs. 8A-8B; FIG. 9). Notably, the same is observed when cells were treated with free fatty acid (FFA) oleic acid (OA) or palmitic acid (PA), respectively (FIGs 7B-7C). Taken together, ADAR-mediated RNA editing of the AUG codon in PNPLA3-148M RNA transcript has the potential to both reduce the protein levels of the disease variant whilst, at the same time, replacing the disease variant with a form of the protein (PNPLA3-148V) that phenocopies in function to that of the wild-type PNPLA3 in liver cells thereby providing an effective treatment for individuals at risk of liver disease or those already suffering from liver disease.

Thus, in a first aspect, the present invention provides a method of preventing or treating liver disease in a subject characterised as having a PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant that phenocopies the function of PNPLA3-148I (wild type).

In a preferred embodiment, the subject is a human subject.

In some embodiments, the method comprises a step of detecting a PNPLA3 rs738409 allele in a sample obtained from the subject prior to administration of the oligonucleotide or the DNA molecule.

In some embodiments, the subject is homozygous for the PNPLA3-148M allele (PNPLA3- 148M/M).

In some embodiments, the subject is heterozygous for the PNPLA3-148M allele (PNPLA3- 148M/I).

In some embodiments, the polynucleotide encoding the PNPLA3-148M protein variant is a messenger ribonucleic acid (mRNA) molecule.

In some embodiments, the codon is converted by ADAR-mediated editing from AUG to IUG

In some embodiments, the oligonucleotide is a guide RNA. In some embodiments, the oligonucleotide is between 30 and 100 nucleotides in length. In some embodiments, the oligonucleotide comprises or consists of an antisense nucleotide sequence at least partially complementary to the region of the polynucleotide encoding the PNPLA3-148M variant spanning codons for amino acid position 148. In some embodiments, the antisense nucleotide sequence is 20-80 nucleotides in length. In some embodiments, the oligonucleotide comprises one or more ADAR-recruiting domains. In some embodiments, the oligonucleotide comprises one or more domains to improve stability. In some embodiments, the DNA molecule is comprised in a viral vector, optionally an adeno- associated viral vector. In some embodiments, the DNA molecule is administered via lipid nanoparticles.

In some embodiments, the polynucleotide encoding the PNPLA3-148M protein variant only undergoes ADAR mediated editing in the cells of the liver. In some embodiments, the lipase activity of PNPLA3 in the liver of the subject is restored to wild-type levels. In some embodiments, the normal function of PNPLA3 in the liver of the subject is restored. In some embodiments of the methods of the disclosure the function of PNPLA3 in the liver of the subject is restored to wild-type levels or a level close to the wild type levels.

In some embodiments, the liver disease is selected from non-alcoholic fatty liver disease (NAFLD), alcoholic-related liver disease (ArLD), liver fibrosis and cirrhosis. In some embodiments, the NAFLD is non-alcoholic steatohepatitis (NASH) or metabolic dysfunction associated steatohepatitis (MASH).

All embodiments described herein in relation to this first aspect of the invention are equally applicable to the further aspects of the invention as described herein below, as appropriate.

In a second aspect, the invention provides a method of editing a target ribonucleic acid (RNA) molecule encoding a PNPLA3-148M protein variant, the method comprising contacting the RNA with an oligonucleotide capable of effecting ADAR-mediated adenosine to inosine alternation of the codon encoding the methionine at position 148, thereby editing the target RNA.

In some embodiments, the codon encoding the methionine at position 148 is converted from AUG to IUG. In some embodiments, the RNA is edited such that the codon for methionine at position 148 is replaced with that for valine.

In a third aspect, the invention provides a method of restoring the function of the PNPLA3 protein in a subject characterised as having the PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding the oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at the codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant. In a fourth aspect, the invention provides use of ADAR-mediated RNA editing to convert a PNPLA3-148M protein variant to a PNPLA3-148V protein variant in subjects having liver disease or at risk of developing liver disease.

In a fifth aspect, the invention provides use of ADAR-mediated RNA editing to prevent or treat liver disease in subjects in need thereof, wherein the subjects are characterized as having a PNPLA3-148M protein variant and the ADAR-mediated RNA editing converts the PNPLA3-148M protein variant to a PNPI-A3-148V protein variant.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 : Schematic depicting the concept underlying the invention - the ADAR-mediated editing of the AUG codon encoding methionine at position 148 of the disease variant PNPLA3-148M to IUG such that the IUG codon is recognized and translated as GUG encoding a valine at position 148. As described herein, the PNPLA3-148V variant can phenocopy the wild-type PNPLA3 protein having an isoleucine at position 148. ADAR will convert the "AUG" to "IUG" and the inosine(l) in the edited codon will be recognized as a guanosine (G) during ribosomal translation.

FIG. 2: DNA sequence of the PNPLA3 gene (Gene ID: 80339; SEQ ID NO: 1).

FIG. 3: mRNA sequence of expressed PNPLA3 gene (SEQ ID NO: 2), where each T=U.

FIG. 4: Amino acid sequence of the PNPLA3 protein, deposited in the Uniprot database as “Q9NST1-PLPL3_HUMAN” (accession number Q9NST1 ; SEQ ID NO: 3).

FIG. 5: Amino acid sequence of the PNPLA3-148M protein variant (SEQ ID NO: 4).

FIG. 6: Amino acid sequence of the PNPLA3-148V protein variant (SEQ ID NO: 5).

FIGs. 7A-7C: Lipid accumulation measured in HuH-7 PNPLA3 KO clonal cells overexpressing PNPLA3 variants with and without free fatty acid treatment as described in Example 1 . HuH-7 cell lines were transduced with BacMam virus expressing PNPLA3-148I, 148M, or 148V at the same volume and MOI (MO 00), respectively for overnight and then incubated without any fatty acid treatment (FIG. 7A), or treatment with 100 pM oleic acid (FIG. 7B), or treatment with 100 pM palmitic acid (FIG. 7C) for 24 hours. After treatment, cells were fixed for neutral lipid staining with Nile red. Stained cells were scanned with a high content imager and lipid content was measured from Nile red intensity. P values were calculated with TTEST. For experiments in HuH-7 cell lines without any fatty acid treatment (FIG. 7A), the statistical significance was also confirmed by an independent statistical analysis accounting for potential sources of variability (data not shown). HuH-7 cells overexpressing PNPLA3-148V and PNPLA3-148I (wild type) had significantly reduced lipid accumulation compared to cells overexpressing PNPLA3-148M. FIGs. 8A-8B: Lipid accumulation measured in primary human hepatocytes (PHH) overexpressing PNPLA3 variants as described in Example 1 . PHH cells from two donors (Hu8356 or HUM4167) were plated and transduced with a PNPLA3 variant BacMam at MOI of 100. After 2 days, cells were fixed for neutral lipid staining with Lipidtox Green. Stained cells were scanned with high content imager. Lipid content was measured for Lipidtox Green intensity and the p values were calculated with TTEST. FIG. 8A: lipid accumulation measured in donor Hu8356; and FIG. 8B: lipid accumulation measured in donor HUM4167. Both PHH donors have the genotype heterozygous for PNPLA3 at amino acid position 148 (i.e., PNPLA3-148M/I).

FIG. 9: Lipid accumulation assay measured in iPS-cell derived hepatocytes as described in Example 1 . At day 5 after plating, FCDI iCell hepatocyte were transduced with a PNPLA3 variant BacMam, at MOI=20, and incubated for another 2 days. The cells were then stained for LipidTox Green (n=4 replicates, and error bars represent the standard error; p value based on one-way ANOVA). The FCDI iCell hepatocyte donor line has the genotype homozygous for the wild type PNPLA3 at amino acid position 148 (i.e., PNPLA3-148I/I).

FIGs. 10A-10B: Lipase activity assay of four purified recombinant full length PNPLA3 fusion protein variants as described in Example 2. FIG. 10A: Michaelis-Menten fit of initial reaction rate versus varying concentrations of fluorescent triglyceride (TG) substrate (18:1-6:0 DNP-C11 TopFluor TG). FIG. 10B: Michaelis-Menten fit of initial reaction rate versus varying concentrations of fluorescent monoglyceride (MG) substrate (heptanoate-MUB).

FIGs. 11A-11 B: Lipase activity assay of four purified recombinant truncated PNPLA3 (1-276) fusion protein variants as described in Example 2. FIG. 11 A: Michaelis-Menten fit of initial reaction rate versus varying concentrations of fluorescent triglyceride (TG) substrate (18:1-6:0 DNP-C1 1 TopFluor TG). FIG. 11 B: Michaelis-Menten fit of initial reaction rate versus varying concentrations of fluorescent monoglyceride (MG) substrate (heptanoate-MUB).

DETAILED DESCRIPTION OF THE INVENTION

A. Definitions

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as is commonly understood by one skilled in the art in the technical field of the invention.

As used herein the term “patatin like phospholipase domain containing 3” or “PNPLA3” refers to a protein product of the PNPLA3 gene present on human chromosome 22, location 22q13.31 (Genbank Gene ID: 80339, FIG. 2, SEQ ID NO: 1). PNPLA3 is expressed primarily in hepatocytes and stellate cells in the liver. PNPLA3 acts to regulate the development of adipocytes and the production and breakdown of fats in hepatocytes and adipocytes. PNPLA3 protein is associated with intracellular lipid droplets, where it has lipase activity and contributes to the regulation of triglycerides. The nucleotide sequence of PNPLA3 mRNA is represented by SEQ ID NO: 2 (FIG. 3).

The protein sequence of human PNPLA3 is deposited in the Uniprot database as “Q9NST1- PLPL3_HUMAN”. The PNPLA3 protein designated the canonical sequence (accession number Q9NST1) consists of 481 amino acids (SEQ ID NO: 3, FIG. 4).

The terms “PNPLA3-I148M”, “PNPLA3-148M” or “PNPLA3-148M disease variant” are used herein interchangeably and refer to a PNPLA3 protein variant that has a methionine residue instead of an isoleucine residue at position 148 of SEQ ID NO: 3 (FIG. 5, SEQ ID NO: 4).

Subjects expressing PNPLA3-148M have the PNPLA3 rs738409 allele, also referred to herein as the “rs738409 allele”. The PNPLA3-148M allele comprises the single nucleotide polymorphism (SNP) rs738409 (C>G), which involves a missense mutation of cytosine to guanine in the codon encoding the amino acid at position 148 of PNPLA3 (SEQ ID NO: 3). Subjects treated in accordance with the methods described herein may be homozygous or heterozygous for the PNPLA3-148M allele.

The SNP rs738409 (C>G) results in the expression of the PNPLA3-148M disease variant, which comprises a substitution of the wild-type amino acid isoleucine to the disease variant amino acid, methionine, at position 148.

PNPLA3-148M is associated with higher risk of liver diseases and increased mortality (Dai et al., 2019; Xiang et al. 2021 ; Unalp-Arida and Ruhl, 2020). PNPLA3-148M has reduced lipase activity compared to wild-type PNPLA3 and causes increased intracellular lipid droplet size in the liver (Huang et al., 2011 ; Dong 2019). In addition, PNPLA3-148M has been shown to have increased stability and enrichment on lipid droplets. This has been shown to suppress adipose triglyceride lipase (ATGL, aka PNPLA2) activity by sequestering the shared co-activating factor CGI-58 (Wang et al., 2019; Dong 2019). These changes both contribute toward increased lipid droplet size and reduced lipolysis in hepatocytes.

The term “PNPLA3-148V” refers to the PNPLA3 protein variant encoded by the edited polynucleotide generated in accordance with the methods of the present invention. ADAR- directed adenosine deamination at the codon encoding amino acid position 148 of SEQ ID NO: 4 results in conversion of adenosine to inosine. Inosine is read as guanine by RNA translational machinery and thus the codon encoding the amino acid at position 148 of SEQ ID NO: 4 is read as GUG and so encodes the amino acid valine (FIG. 6, SEQ ID NO: 5). As demonstrated herein, expression of PNPLA3-148V restores function of the PNPLA3 protein. In particular, PNPLA3- 148V restores lipase function in liver cells. Thus, PNPLA3-148V overexpressed in hepatic cells (HuH7, primary hepatocytes, iPS-cell derived hepatocytes) reduces intracellular lipid accumulation compared to cells overexpressing PNPLA3-148M and restores lipid accumulation to a level close to cells overexpressing wild-type PNPLA3-148I (FIGs. 7A-7C, FIGs. 8A-8B, FIG. 9).

The term “ADAR” or “Adenosine deaminases acting on RNA”, refers to a class of enzymes that carry out site-directed conversion of adenosine to inosine in double-stranded RNA (dsRNA). ADAR enzymes contain multiple dsRNA binding domains with which they bind to dsRNA and a deaminase domain having catalytic activity. ADARs bind to dsRNA and detect an A-C mismatch between the two RNA molecules. The deaminase domain then converts the adenosine of the mismatch to inosine (Booth et al., 2023; Bellingrath et al., 2023).

The term “ADAR editing” refers to the editing of RNA molecules carried out by ADAR enzymes. ADAR editing may also be referred to as “ADAR-mediated editing”, “ADAR-mediated adenosine to inosine alteration” or “ADAR-directed adenosine deamination”. ADAR editing may be an endogenous process carried out by endogenous ADARs. Alternatively, ADAR enzymes may be exogenously expressed so as to employ RNA editing as a gene editing tool in cells that lack or have low levels of expression of endogenous ADAR enzymes.

The term “oligonucleotide” is used herein to refer to the molecule capable of effecting ADAR editing in the context of the present invention. The term oligonucleotide is used herein broadly to refer to any oligonucleotide molecule capable of hybridising to the target sequence within the polynucleotide encoding the PNPLA3-148M variant and effecting ADAR editing. The oligonucleotide may comprise single- and double-stranded regions and adopt any suitable conformation as described elsewhere herein. The oligonucleotide for use in accordance with the methods described herein may also be referred to as a “guide oligonucleotide” i.e., an oligonucleotide capable of effecting ADAR-mediated adenosine to inosine alteration. As described elsewhere herein, a guide oligonucleotide is at least partially complementary to the target PNPLA3-148M polynucleotide. The guide oligonucleotide may be 30 to 100 nucleotides in length. The guide oligonucleotide may comprise an antisense region and one or more ADAR- recruiting domains.

The term “antisense region” refers to a region of the guide oligonucleotide that is complementary to a target polynucleotide which comprises a target adenosine. The antisense region may also be referred to as the antisense nucleotide sequence. The binding of the antisense region to the target polynucleotide produces a double stranded nucleic acid structure that serves as a substrate for ADAR editing. In some embodiments, the double stranded nucleic structure that serves as a substrate for ADAR editing is a dsRNA. The dsRNA structure may also facilitate recruitment of ADAR. The antisense region of the oligonucleotides described herein may be 20 to 80 nucleotides in length.

As used herein, the term “at least partially complementary” means that the oligonucleotide is capable of hybridising to at least a portion of the target polynucleotide, with one or more mismatches permitted between the antisense nucleotide sequence and target polynucleotide. For instance, the antisense nucleotide sequence may be at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% complementary to the target PNPLA3-148M polynucleotide.

The term “ADAR recruiting domain” refers to a region or domain of the oligonucleotide that facilitates recruitment and binding of ADAR enzymes. The ADAR recruiting domain may have a specific sequence or other structural feature that confers the ability to recruit ADAR. The ADAR recruiting domain may also increase on-site editing specificity and efficiency and reduce bystander and transcriptome-wide off-target editing.

The terms “nucleic acid” and “polynucleotide” are used herein interchangeably and refer generally to any DNA or RNA molecule, either single- or double-stranded and, if single-stranded, the molecule of its complementary sequence. In discussing nucleic acid molecules, a sequence or structure of a particular nucleic acid molecule may be described herein according to the normal convention of providing the sequence in the 5' to 3' direction. In some embodiments of the invention, nucleic acids or polynucleotides are "isolated." This term, when applied to a nucleic acid molecule, refers to a nucleic acid molecule that is separated from sequences with which it is immediately contiguous in the naturally occurring genome of the organism in which it originated. For example, an "isolated nucleic acid" may comprise a DNA molecule inserted into a vector, such as a plasmid or virus vector, or integrated into the genomic DNA of a prokaryotic or eukaryotic cell or non-human host organism. When applied to RNA, the term "isolated polynucleotide" refers primarily to an RNA molecule encoded by an isolated DNA molecule as defined above. Alternatively, the term may refer to an RNA molecule that has been purified/separated from other nucleic acids with which it would be associated in its natural state (i.e., in cells or tissues). An isolated polynucleotide (either DNA or RNA) may further represent a molecule produced directly by biological or synthetic means and separated from other components present during its production. The term “liver disease” encompasses pathological conditions and disorders associated with the liver. In the context of the present invention “liver disease” encompasses non-alcoholic fatty liver disease (NAFLD), non-alcoholic steatohepatitis (NASH), alcoholic-related liver disease (ArLD), liver fibrosis and cirrhosis, as well as HBV/HCV progression, hepatocellular carcinoma and liver transplant. In some embodiments of the methods described herein, the liver disease is nonalcoholic steatohepatitis (NASH). NAFLD has been renamed to “metabolic dysfunction associated steatotic liver disease” or “MAFLD.” Likewise, NASH has been renamed to “metabolic dysfunction associated steatohepatitis” or “MASH.” NAFLD and MAFLD are used interchangably herein. NASH and MASH are used interchangeably herein.

The terms “non-alcoholic steatohepatitis” or “NASH” and “metabolic dysfunction associated steatohepatitis” or “MASH” refer to a disorder of the liver caused by the excess buildup of lipid along with fibrosis and inflammation. NASH is a more clinically serious form of NAFLD and is characterised by hepatocyte damage, hepatocyte swelling, inflammatory infiltrate, liver inflammation and the early development of perisinusoidal fibrosis. The term non-alcoholic steatohepatitis encompasses pre-cirrhotic non-alcoholic steatohepatitis (stage F2/F3).

The terms “individual”, “subject” and “patient” are used herein interchangeably to refer to an animal or human. In some embodiments, the subject is a mammal, such as a primate. In other embodiments, the subject is a human.

The terms “prevent”, “preventing” and “prevention” refer to the prophylactic treatment of a subject without the specified condition, potentially a subject at risk of developing the specified condition. The methods of prevention described herein involve administration to a subject, an oligonucleotide capable of effecting ADAR editing, or a DNA molecule encoding an oligonucleotide capable of effecting ADAR editing, so as to achieve one or more of the following effects: fully, or at least partially, protect the subject from the onset, or the symptomatic onset, of the specified condition i.e., liver disease, particularly NASH or MASH.

The terms “treat”, “treating” and “treatment” refer to the therapeutic treatment of a subject with the specified condition i.e., liver disease, particularly NASH or MASH. In the broadest application of the present invention, “treating” or “treatment” refers to the treatment of liver disease. The methods of “treatment” described herein involve administration to a subject, an oligonucleotide capable of effecting ADAR editing, or a DNA molecule encoding an oligonucleotide capable of effecting ADAR editing, so as achieve one or more of the following effects: ameliorating or stabilising the specified condition, reducing or eliminating the symptoms of the condition and/or slowing or eliminating the progression of the condition. Treatment may provide a cure for the specified condition or at least prolong the survival of a subject beyond that expected in the absence of such treatment.

Treating and/or preventing may include one or more of the following: (i) delaying the onset of; (ii) slowing/halting progression; (iii) reducing the frequency of; and/or (iv) reducing the severity of, one or more or all of the symptoms of the specified condition in a subject relative to a subject which does not receive the specified treatment.

The term “effective amount” refers to the quantity of the oligonucleotide or the DNA molecule encoding the oligonucleotide which will elicit the desired biological response (ADAR-mediated adenosine to inosine alteration of a polynucleotide encoding the PNPLA3-148M protein variant and treatment or prevention of liver disease) in an animal or human subject. The amount deemed to be an “effective amount” may vary depending on the oligonucleotide, the stage and/or severity of the disease, and/or the age and/or weight of the subject to be treated. An effective amount can be readily determined by one skilled in the art, for example a physician treating a subject with liver disease.

The “effective amount” is either a prophylactically effective amount or a therapeutically effective amount depending on whether the method is for the prevention or treatment of liver disease. A prophylactically effective amount is typically an amount of the oligonucleotide or the DNA molecule encoding the oligonucleotide required to prevent and/or delay the onset of symptoms characterising the disease. A therapeutically effective amount is typically an amount of the oligonucleotide or the DNA molecule encoding the oligonucleotide effective to ameliorate or reduce one or more symptoms of liver disease and/or the amount effective to cure the disease.

As used in the specification and claims, the singular form “a”, “an” and “the” includes plural references unless the context clearly dictates otherwise.

B. Methods of preventin'] or treating liver disease

The present invention is directed to methods of preventing or treating liver disease in subjects, particularly humans, characterised as having the PNPLA3-148M protein variant. As described elsewhere herein, the PNPLA3-148M protein variant differs from “wild-type” human PNPLA3 by virtue of a methionine at position 148 of the protein instead of an isoleucine residue. The PNPLA3-148M protein variant has decreased lipase activity and has been linked to the development and progression of liver disease. The methods of the invention are based on editing a polynucleotide encoding the PNPLA3-148M variant such that the codon encoding the methionine at position 148 is altered. The edited polynucleotide encodes a PNPLA3 protein variant having valine at position 148 instead of methionine. This protein variant is referred to herein as the PNPLA3-148V protein variant. As reported herein, PNPLA3-148V reduces lipid content in cells compared to PNPLA3-148M levels, effectively restoring lipid content in liver cells to levels similar to wild-type. The methods of the invention are thus effective in restoring PNPLA3 function as a means to prevent or treat liver disease.

It follows, that in a first aspect the present invention provides a method of preventing or treating liver disease in a subject characterised as having a PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

In a further aspect, the invention provides a method of restoring the function of the PNPLA3 protein in a subject characterised as having the PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding the oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

Subjects to be treated

Subjects to be treated in accordance with the methods described herein are characterised as having a PNPLA3-148M protein variant. In preferred embodiments, the subject to be treated is a human subject. As detailed elsewhere herein, the PNPLA3-148M protein variant is associated with the development of liver disease. It follows that subjects characterised as having the PNPLA3-148M variant are at greater risk of liver disease as compared with the average risk across the general population. The methods described herein are thus targeted to the group of subjects characterised by the presence of this disease variant.

In some embodiments, the methods of the present invention are for the prevention of liver disease in a subject. For such embodiments, the subject may be an individual identified as “at risk” of developing liver disease, having not yet developed symptoms of liver disease. Subjects at risk may be characterised as having the risk allele rs738409 C>G, also referred to herein as the PNPLA3 rs738409 allele. Genome wide association studies have shown rs738409 C>G to be one of the strongest genetic markers for liver disease, and in particular nonalcoholic steatohepatitis (NASH) or metabolic dysfunction associated steatohepatitis (MASH). Subjects at risk may be homozygous or heterozygous for the PNPLA3-148M allele. In preferred embodiments, the subject to be treated in accordance with the methods described herein is homozygous for the PNPLA3 rs738409 allele i.e., the risk allele.

In some embodiments, the methods of the present invention are for the treatment of liver disease in a subject. For such embodiments, the subject will have already been diagnosed as having liver disease. For instance, the subject may have been newly-diagnosed as having liver disease. A subject that is newly-diagnosed as having liver disease may be in the early stages of liver disease. A subject diagnosed as having liver disease and treated in accordance with the methods described herein may have pre-cirrhotic NASH (F2/F3). Alternatively, a subject that is newly-diagnosed as having liver disease may have advanced-stage liver disease. Accordingly, the methods of the present invention may both treat existing liver disease in the subject and also prevent the development of further symptoms and advancement of the pathology of the liver disease.

Subjects for treatment according to the methods of the present invention may have been previously treated for liver disease and/or be receiving concomitant therapy. In particular, subjects for treatment may have previously received or currently be receiving treatment for nonalcoholic fatty liver disease (NAFLD), non-alcoholic steatohepatitis (NASH), metabolic dysfunction associated steatohepatitis (MASH) alcoholic-related liver disease, liver fibrosis or cirrhosis, or diabetes. Standard of care treatments for these disease indications are known to those skilled in the art and could be used alongside the methods described herein.

The methods of the invention may include a step of testing a sample obtained from a subject for the presence of the PNPLA3-148M protein variant. This step will typically precede the administration of the oligonucleotide or the DNA molecule encoding the same. In some embodiments, the methods include a step of testing a sample obtained from a subject for the presence of the PNPLA3 rs738409 allele. In particular, the step of testing a sample may determine that the subject is at risk of developing liver disease, particularly in a subject that is not yet symptomatic for disease. The step of testing the sample obtained from the subject for the presence of the PNPLA3 rs738409 allele may comprise testing the sample for mRNA encoding the PNPLA3-148M variant. Alternatively, or in addition, the step of testing the sample may comprise testing the genomic DNA within the sample for the presence of one or more copies of the PNPLA3 rs738409 allele, particularly the rs738409 C>G SNP.

The sample obtained from the subject for use in accordance with the methods described herein may be any suitable biological sample known or suspected to contain genomic DNA or mRNA encoding PNPLA3. The skilled person will appreciate how to select an appropriate sample to detect the gene or mRNA of interest in or at a relevant location. The sample may be obtained, for example, from liver tissue, hepatocytes, blood, and/or a blood processing product, such as plasma or serum. The sample may comprise extracted RNA.

Samples may be tested for DNA or mRNA of the PNPLA3 rs738409 allele using methods known in the art. For example, the level of expression of mRNA in a sample may be determined by detecting a transcribed polynucleotide, or portion thereof, e.g. PNPLA3-148M mRNA. RNA may be extracted from cells using RNA extraction techniques including, for example, using acid phenol/guanidine isothiocyanate extraction (RNAzol B; Biogenesis), RNEASY™ RNA preparation kits (Qiagen) or PAXgene (PreAnalytix, Switzerland). Typical assay formats utilizing ribonucleic acid hybridization include nuclear run-on assays, RT-PCR, Rnase protection assays, northern blotting, in situ hybridization, and microarray analysis.

The presence of a PNPLA3 rs738409 allele may be determined using a nucleic acid probe. The term “probe,” as used herein, refers to any molecule that is capable of selectively binding to a specific sequence, e.g. to an mRNA or polypeptide. Probes can be synthesized by one of skill in the art or derived from appropriate biological preparations. Probes may be specifically designed to be labelled. Examples of molecules that can be utilized as probes include, but are not limited to, RNA, DNA, proteins, antibodies, and organic molecules.

The presence of the PNPLA3 rs738409 allele might be detected via the rs738409 (C>G) SNP. Samples may be tested for the PNPLA3 rs738409 allele using techniques known in the art. The PNPLA3 rs738409 allele may be detected using genotyping techniques including sequencing, quantitative PCR, chip detection, refractory mutation system PCR, competitive allele-specific PCR, or other such techniques.

ADAR-mediated editing

The present methods employ the technique of adenosine deaminase acting on RNA (ADAR)- mediated editing to alter the codon encoding the methionine at position 148 of the PNPLA3- 148M disease variant. ADAR-mediated editing converts adenosine residues to inosine residues and in the context of the present invention, the AUG codon encoding the methionine at position 148 is converted to an IUG codon, encoding a valine residue. As described elsewhere herein, ADAR-mediated editing is a powerful tool for the alteration of ribonucleic acid (RNA) molecules. This technique possesses significant advantages over other gene editing techniques, particularly DNA editing techniques. RNA editing as opposed to DNA editing can be transient in nature for the life of the edited RNA molecule. Editing of RNA as opposed to permanent editing of DNA may be more acceptable in a therapeutic setting since the effect can easily be reversed by ceasing administration of the therapeutic oligonucleotide to the patient. In contrast, safety concerns persist with DNA editing techniques such as CRISPR-Cas due to the permanent alterations at the genomic level and the oncogenic risk associated therewith.

Subjects to be treated in accordance with the methods described herein are administered an oligonucleotide or a DNA molecule encoding an oligonucleotide. The oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M variant, specifically editing of the codon encoding the methionine at position 148. In some embodiments, the polynucleotide encoding the PNPLA3-148M protein variant is a messenger ribonucleic acid (mRNA) molecule.

Several different approaches are known in the art for designing oligonucleotides capable of effecting ADAR-mediated editing of polynucleotides or target RNAs. Oligonucleotides for use in the methods described herein may be designed to effect ADAR-mediated editing of polynucleotides or target RNAs as described previously. See, e.g., Bellingrath et al. 2023; Reautschnig et al. 2022; Katrekar et al. 2022; Katrekar et al. 2019; Monian et al. 2022; Montiel- Gonzalez et al. 2019; Yi et al. 2022.

In the context of the present invention, the oligonucleotide may be capable of effecting ADAR- mediated editing via the recruitment of endogenous ADAR enzymes. Alternatively or in addition, the oligonucleotide may be capable of effecting ADAR-mediated editing via the recruitment of an exogenous ADAR enzyme present within the cells of the subject.

In preferred embodiments, the oligonucleotide is a guide RNA (gRNA), specifically a gRNA designed to target the region of the PNPLA3 polynucleotide spanning the codon encoding the methionine at position 148. In certain embodiments, the gRNA is between 30 and 100 nucleotides in length.

The oligonucleotide (e.g. gRNA) must at least partially hybridise with the polynucleotide (e.g. mRNA) encoding the PNPLA3-148M protein variant, thereby forming a double stranded RNA (dsRNA) molecule. Thus, hybridisation by the oligonucleotide (e.g. gRNA) forms a dsRNA molecule capable of recruiting one or more ADAR enzymes for ADAR-mediated editing.

The oligonucleotide (e.g. gRNA) may, thus, comprise or consist of an antisense nucleotide sequence (also referred to herein as an antisense region) at least partially complementary to the region of the PNPLA3-148M mRNA spanning position 148. In certain embodiments, the antisense nucleotide sequence is 20-80 nucleotides in length. As used herein, the term “at least partially complementary” means that the antisense nucleotide sequence is capable of hybridising to its target polynucleotide to form a double stranded nucleic acid structure (e.g. a dsRNA), but that there may be one or more mismatches between the antisense nucleotide sequence and target polynucleotide. For instance, the antisense nucleotide sequence may be at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% complementary to the target region of the PNPLA3-148M polynucleotide (e.g. mRNA).

The antisense nucleotide sequence may form an A-C mismatch at the site of the target polynucleotide to be edited. In other words, the antisense nucleotide sequence of the oligonucleotide may comprise a cytosine at the position opposite to the target adenosine of the target polynucleotide. The A-C mismatch is detected by ADAR enzymes and this mismatch directs editing of the adenosine residue of the target polynucleotide to inosine.

In some embodiments, the antisense nucleotide sequence comprises further mismatches, in addition to the A-C mismatch at the position opposite to the target adenosine residue of the target polynucleotide. For example, across the full region of complementarity between the antisense region of the oligonucleotide and the corresponding region of the target polynucleotide, there may be 1 , 2, 3, 4, 5, 6, 7, 8, 9 or 10 nucleotide mismatches.

ADAR enzymes preferentially deaminate adenosines in an A-C mismatch over those that occur in an A-A or A-G mismatch, or in an A-U pairing (Bellingrath et al. 2023). However, ADAR enzymes may carry out bystander editing at off-target adenosines, including adenosines involved in base pairing in the double stranded nucleic acid structure. The preferential targeting of A-C mismatches compared to other mismatches enables targeted modifications to reduce bystander editing in adenosines vulnerable to bystander edits. Thus, in some embodiments, the oligonucleotide may comprise A-G mismatches at non-target adenosines that are vulnerable to bystander edits.

In certain embodiments, there may be discontinuous segments of complementary nucleotide sequences between the oligonucleotide (e.g. gRNA) and the polynucleotide target. Each segment of complementary nucleotide sequences may be 100% complementary or may be partially complementary with the proviso that there must be at least one mismatch at the target adenosine site. Discontinuous complementary segments of the oligonucleotide (e.g. gRNA) and the polynucleotide target may be separated by groups of non-complementary nucleotides. Such discontinuous hybridisation may increase target specificity.

In certain embodiments, the oligonucleotide (e.g. gRNA) and/or the antisense region comprises one or more intramolecular secondary structures. In some embodiments, the oligonucleotide (e.g. gRNA) and/or the antisense nucleotide sequence comprises internal loops. Such loops may reduce bystander editing of the cell’s RNA (Booth et al. 2023).

In certain embodiments, the oligonucleotide may be chemically modified so as to improve stability, specificity and/or efficiency. Suitable chemical modifications are known in the art and could be readily employed by the skilled person implementing the current invention.

Chemical modifications may be particularly important for oligonucleotides that are to be administered directly to the subject (rather than encoded by DNA molecules). In preferred embodiments, the oligonucleotide is a guide RNA having one or more chemical modifications.

In some embodiments, the oligonucleotides comprise one or more chemical modifications of the ribose sugar and/or phosphate backbone.

In some embodiments, the oligonucleotides comprise one or more chemical modifications selected from 2’-O-methyl modified ribose groups (2'-O-Me); 2’-fluoro modified ribose groups (2’-F); 2’-0-methoxyethyl modified ribose groups (2’-MOE); locked nucleic acid modified nucleotide (LNA); and constrained ethyl modified nucleotide (cEt).

In some embodiments, the oligonucleotides comprise one or more chemical modifications selected from phosphorothioate backbones (PS) and phosphoramidate backbones (PN).

In some embodiments, the oligonucleotides comprise one or more chemical modifications selected from: 2’-O-methyl modified ribose groups (2'-O-Me); and phosphorothiate backbones (PS).

2’-0-methylation refers to the substitution of a 2’-OH group in a ribose sugar of the ribosephosphate back bone with a 2’-O-methyl group. 2’-0-methylation may protect the oligonucleotide from nuclease degradation and increase thermal stability (Adachi et al. 2021). Phosphorothiate backbones refer to oligonucleotide backbones wherein an oxygen atom in the phosphodiester group is substituted with a sulphur atom. Modifications to introduce phosphorothiate backbones may increase bioavailability and cellular uptake of the oligonucleotides (Adachi et al. 2021).

In some embodiments, the oligonucleotides comprise N-acetylgalactosamine (GalNAc). For example, the oligonucleotide can be covalently conjugated to GalNAc at the 5’-end or the 3’ of the oligonucleotide, optionally via a linker. GalNAc is a sugar group that recognizes and binds to the asialoglycoprotein receptor (ASGPR). ASGPR is a cell surface protein expressed on the surface of hepatocytes. Thus, modifying the oligonucleotide with GalNAc may allow efficient uptake of the oligonucleotide into hepatocytes. In some embodiments, the oligonucleotides are covalently conjugated to a triantennary GalNAc ligand at the 5’-end or the 3’-end of the oligonucleotide, optionally via a linker.

In certain embodiments, the oligonucleotide may have one or more additional domains beyond the antisense region. For example, the oligonucleotide may have ADAR recruiting domains (ARDs) to recruit endogenous ADARs and/or domains to improve stability of the molecule.

The oligonucleotide may have exonuclease-resistant structures at the 5’ and/or 3’ ends of the molecule. For instance, the oligonucleotide may be chemically modified to include a 5’ cap and a 3’ poly(A)tail, to improve resistance to exonuclease degradation.

The oligonucleotide may be linear or circular. For instance, circularization of the oligonucleotide may confer greater resistance to exonuclease degradation (Booth et al. 2023). In some embodiments, the oligonucleotide is a circular RNA. A circular RNA can be chemically synthesized or produced in cellular systems followed by purification of the circular RNA, such as described in Yi et al. (2022). In some embodiments, a DNA molecule (e.g., viral vector) encodes the oligonucleotide, wherein the oligonucleotide is a genetically encoded circular RNA, such as described in Katrekar et al. (2022).

The oligonucleotide may comprise or consist of one or more intramolecular secondary structure features. In some embodiments, the oligonucleotide comprises one or more hairpin loops. Hairpin loop structures may assist recruitment of ADAR enzymes to the target dsRNA complex.

Formulation and routes of administration

The methods of the present invention involve a step of administering an oligonucleotide capable of effecting ADAR-mediated editing or administering a DNA molecule encoding the same. It follows, that in some embodiments, the oligonucleotide is formulated for direct administration to the subject, with or without conjugation with GalNAc. In alternative embodiments, the oligonucleotide is encoded by a DNA molecule that must be formulated for administration to the subject.

In this regard, there is a requirement for the oligonucleotide or the DNA molecule encoding the oligonucleotide to be delivered to the liver cells so that the oligonucleotide comes into contact with its target polynucleotide in the desired location within the body.

The oligonucleotide or DNA molecule may be synthesized and formulated for delivery in accordance with any suitable method known to those skilled in the art. In particular, the oligonucleotide or DNA molecule may be synthesized by solution-phase or solid phase synthesis.

For embodiments wherein the oligonucleotide is to be administered directly to the subject, the oligonucleotide may be chemically synthesised or in vitro transcribed prior to administration. As described above, the oligonucleotide may be chemically modified to improve stability.

For embodiments wherein a DNA molecule encoding an oligonucleotide is administered to the subject, the DNA molecule may be any suitable plasmid or viral vector. In some embodiments, the DNA molecule is comprised in a viral vector, optionally an adenoviral vector or adeno- associated viral vector. In some embodiments, the DNA molecule is administered via lipid nanoparticles.

The oligonucleotides or DNA molecules described herein may be administered to the subject via any suitable route of administration. In some embodiments, the oligonucleotide or DNA molecule is administered via intravenous administration. In some embodiments, the oligonucleotide or DNA molecule is administered via subcutaneous administration. Systemic injection of adeno- associated virus (AAV) results in high transduction of the liver and as such, delivery via AAV vector may be a particularly suitable route of administration for the present methods especially when the therapeutic nucleotide is long (>100 bases) and/or for encoding a circular ADAR recruiting RNA (circ-adRNA). Alternatively, the oligonucleotide or DNA molecule may be locally administered by direct injection to the site. In particular, direct administration may cause the PNPLA3-148M protein variant to undergo ADAR-mediated RNA editing in the cells of the liver only.

The oligonucleotide or DNA molecule may be delivered using drug delivery systems known to those skilled in the art. In particular, the oligonucleotide may be delivered using a nanoparticle, a lipid nanoparticle, a polyplex nanoparticle, a lipoplex nanoparticle, a dendrimer, a polymer, liposomes, or a cationic delivery system. Positively charged cationic delivery systems facilitate binding of negatively charged oligonucleotides and enable efficient uptake of the oligonucleotide by cells through the negatively charged cell membrane. Alternatively, the oligonucleotide may be delivered in vesicles or micelles formed by cationic lipids, dendrimers or polymers. Administration with vesicles or micelles may prevent degradation of the oligonucleotide during systemic administration.

In some embodiments, particularly wherein the oligonucleotide is formulated for direct administration, the oligonucleotide may be conjugated with one or more molecules of N- acetylgalactosamine (GalNAc). GalNAc is a sugar group that recognizes and binds to the asialoglycoprotein receptor (ASGPR). ASGPR is a cell surface protein expressed on the surface of hepatocytes. Thus, conjugation of the oligonucleotide to GalNAc should facilitate efficient uptake of the oligonucleotide into hepatocytes.

Prevention /treatment of liver disease

The methods in accordance with the present invention are for the prevention and/or treatment of liver disease. It follows, that the polynucleotide encoding the PNPLA3-148M protein variant must undergo ADAR-mediated RNA editing in the cells of the liver. In some embodiments, the polynucleotide encoding the PNPLA3-148M protein variant only undergoes ADAR-mediated RNA editing in the cells of the liver. In preferred embodiments, the lipase activity of the PNPLA3 protein in liver cells is restored to wild-type levels as a result of the methods described herein. In other embodiments, as a result of the methods described herein PNPLA3-148V attenuates the loss of catalytic activity of the PNPLA3-148M variant and restores activity of the PNPLA3 protein to close to that of wild-type PNPLA3. Such a 148M-to-148V conversion will also attenuate the gain of function phenotypes associated with the PNPLA3-148M variant.

The methods of the present invention may be used to prevent or treat any form of liver disease associated with the PNPLA3-148M disease variant. The terms “prevent” and “treat” are to be understood as defined elsewhere herein. In some embodiments, the methods are for the prevention or treatment of a liver disease selected from the group consisting of: non-alcoholic fatty liver disease (NAFLD); alcoholic-related liver disease; liver fibrosis; and cirrhosis. In preferred embodiments, the methods are for the prevention or treatment of non-alcoholic steatohepatitis (NASH) or metabolic dysfunction associated steatohepatitis (MASH). In certain embodiments, the methods are for the treatment of subjects having pre-cirrhotic NASH or MASH.

C. Uses In a further aspect, the invention provides use of ADAR-mediated RNA editing to convert a PNPLA3-148M protein variant to a PNPLA3-148V protein variant in subjects having liver disease or at risk of developing liver disease. In a particular embodiment, the invention provides use of ADAR-mediated RNA editing to convert a PNPLA3-148M protein variant to a PNPLA3-148V protein variant in subjects having non-alcoholic steatohepatitis (NASH) or metabolic dysfunction associated steatohepatitis (MASH), or at risk of developing NASH or MASH.

In a yet further aspect, the present invention provides use of ADAR-mediated RNA editing to prevent or treat liver disease in subjects in need thereof, wherein the subjects are characterized as having a PNPLA3-148M protein variant and the ADAR-mediated RNA editing converts the PNPLA3-148M protein variant to a PNPLA3-148V protein variant. In a particular embodiment, the invention provides use of ADAR-mediated RNA editing to prevent or treat NASH or MASH in subjects in need thereof, wherein the subjects are characterized as having a PNPLA3-148M protein variant and the ADAR-mediated RNA editing converts the PNPLA3-148M protein variant to a PNPLA3-148V protein variant.

All embodiments described above in connection with the preceding aspects of the invention are equally applicable to the uses described herein.

References

Adachi et al. (2021), “From Antisense RNA to RNA Modification: Therapeutic Potential of RNA- Based Technologies”, Biomedicines, 9 (550), 1-26

Bellingrath et al. (2023), “Programmable RNA editing with endogenous ADAR enzymes - a feasible option for the treatment of inherited retinal disease?”, Frontiers in Molecular Neuroscience, 24 (16), 1092913

Booth et al. (2023), “RNA editing: Expanding the potential of RNA therapeutics”, Molecular Therapy, 31 (6), 1533-1549

Dai et al., (2019), “Association between PNPLA3 rs738409 polymorphism and nonalcoholic fatty liver disease (NAFLD) susceptibility and severity A meta-analysis”, (2019) 98:7(e14324)

Dong (2019), “PNPLA3-A potential therapeutic target for personalized treatment of chronic liver disease”, Front Med (Lausanne) 2019; 6:304.

Fan et al. (2016) “PNPLA3 rs738409 Polymorphism Associated with Hepatic Steatosis and Advanced Fibrosis in Patients with Chronic Hepatitis C Virus: A Meta-Analysis” Gut and Liver 2016, 10(3) 456-463 Fornwald et al. (2016), “Gene Expression in Mammalian Cells Using BacMam, a Modified Baculovirus System”, Methods in Molecular Biology, 1350, 95-116. DOI:10.1007/978-1 -4939- 3043 -2_5

Hsueh et al. (2022), “Impact of PNPLA3 p.l148M and Hepatic Steatosis on Long-Term Outcomes for Hepatocellular Carcinoma and HBsAg Seroclearance in Chronic Hepatitis B” Journal of Hepatocellular Carcinoma, 2022, 9, 301 -313

Huang et al., (201 1), “Expression and characterization of a PNPLA3 protein isoform (I148M) associated with nonalcoholic fatty liver disease”, J Biol Chem. 2011 Oct 28;286(43):37085-93.

Katrekar et al. (2022), “Efficient in vitro and in vivo RNA editing via recruitment of endogenous ADARs using circular guide RNAs” Nature Biotech. 2022 (40), 938-945

Katrekar et al. (2019), “IN vivo RNA editing of point mutations via RNA-guided adenosine deaminases” Nature Methods 2019 (16), 239-242

Kolla et al. (2018), “PNPLA3 Association with Alcoholic Liver Disease in a Cohort of Heavy Drinkers”, Alcohol and Alcoholism 2018, 53(4), 357-360

Linden et al. (2019), “Pnpla3 silencing with antisense oligonucleotides ameliorates nonalcoholic steatohepatitis and fibrosis in Pnpla3 I148M knock-in mice”, Molecular metabolism, 22, 49-61

Monian et al. (2022), “Endogenous ADAR-mediated RNA editing in non-human primates using stereopure chemically modified oligonucleotides” Nature Biotech. 2022 (40), 1093-1102

Montiel-Gonzalez et al. (2019) “Current strategies for Site-Directed RNA Editing using ADARs” Methods 2019 (156), 16-24

Pingitore and Romeo (2019), “The role of PNPLA3 in health and disease”, Biochim Biophys Acta Mol Cell Biol Lipids. 2019 Jun;1864(6):900-906

Reautschnig et al. (2022), “CLUSTER guide RNAs enable precise and efficient RNA editing with endogenous ADAR enzymes in vivo” Nature Biotech. 2022 (40), 759-768.

Salari et al. (2021), “Association between PNPLA3 rs738409 polymorphism and nonalcoholic fatty liver disease: a systematic review and meta-analysis”, BMC Endocrine Disorders, 21 (125), 1-12

Stickel at al. (2015), “PNPLA3 Genetic Variation in Alcoholic Steatosis and Liver Disease Progression”, Hepatobiliary Surgery and Nutrition, 2015, 4(3) 152-160 Unalp-Arida and Ruhl (2020), “Patatin-Like Phospholipase Domain-Containing Protein 3 I148M and Liver Fat and Fibrosis Scores Predict Liver Disease Mortality in the U.S. Population”, Hepatology, 71 (3), 820-834

Wang et al. (2019), “PNPLA3, CGI-58, and inhibition of hepatic triglyceride hydrolysis in mice”, Hepatology 2019 Jun;69(6):2427-2441 .

Xiang et al. (2021), “Research progress, challenges and perspectives on PNPLA3 and its variants in Liver Diseases”, Journal of Cancer, 12 (19), 5929-5937

Yi et al. (2022), “Engineered circular ADAR-recruiting RNAs increase the efficiency and fidelity of RNA editing in vitro and in vivo” Nature Biotech. 2022 (40), 946-955.

Reference

All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

EMBODIMENTS

Embodiment 1 is a method of preventing or treating liver disease in a subject characterised as having a PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR- mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

Embodiment 2 is the method of embodiment 1 , wherein the subject is a human subject.

Embodiment 3 is the method of embodiment 1 or embodiment 2, wherein prior to administration of the oligonucleotide or the DNA molecule, the method comprises a step of detecting a PNPLA3 rs738409 allele in a sample obtained from the subject.

Embodiment 4 is the method of any one of embodiments 1-3, wherein the subject is homozygous, or heterozygous, for the PNPLA3 rs738409 allele. Embodiment 5 is the method of any one of embodiments 1-4, wherein the polynucleotide encoding the PNPLA3-148M protein variant is a messenger ribonucleic acid (mRNA) molecule.

Embodiment 6 is the method of any one of embodiments 1-5, wherein the codon is converted by ADAR-mediated editing from AUG to IUG.

Embodiment 7 is the method of any one of embodiments 1-6, wherein the oligonucleotide is a guide RNA.

Embodiment 8 is the method of any one of embodiments 1-7, wherein the oligonucleotide is between 30 and 100 nucleotides in length.

Embodiment 9 is the method of any one of embodiments 1-8, wherein the oligonucleotide comprises or consists of an antisense nucleotide sequence at least partially complementary to the region of the polynucleotide encoding the PNPLA3-148M variant spanning position 148.

Embodiment 10 is the method of embodiment 9, wherein the antisense nucleotide sequence is 20-80 nucleotides in length.

Embodiment 11 is the method of any one of embodiments 1 -10, wherein the oligonucleotide comprises one or more ADAR-recruiting domains.

Embodiment 12 is the method of any one of embodiments 1 -11 , wherein the oligonucleotide comprises one or more domains to improve stability.

Embodiment 13 is the method of any one of embodiments 1 -12, wherein the DNA molecule is comprised in a viral vector, optionally an adeno-associated viral vector.

Embodiment 14 is the method of any one of embodiments 1 -13, wherein the DNA molecule is administered via lipid nanoparticles.

Embodiment 15 is the method of any one of embodiments 1 -14, wherein the polynucleotide encoding the PNPLA3-148M protein variant only undergoes ADAR-mediated editing in the cells of the liver.

Embodiment 16 is the method of any one of embodiments 1-15, wherein the lipase activity of PNPLA3 in the liver of the subject is increased relative to an initial lipase activity level and/or restored to wild-type levels. Embodiment 17 is the method of any one of embodiments 1 -16, wherein the liver disease is selected from non-alcoholic fatty liver disease (NAFLD), alcoholic-related liver disease (ArLD), liver fibrosis and cirrhosis.

Embodiment 18 is the method of embodiment 17, wherein the NAFLD is nonalcoholic steatohepatitis (NASH) or metabolic dysfunction associated steatohepatitis (MASH).

Embodiment 19 is a method of editing a target ribonucleic acid (RNA) molecule encoding a PNPLA3-148M protein variant, the method comprising contacting the target RNA with an oligonucleotide capable of effecting ADAR-mediated adenosine to inosine alteration of the codon encoding the methionine at position 148, thereby editing the target RNA.

Embodiment 20 is the method of embodiment 19, wherein the codon encoding the methionine at position 148 is converted from AUG to IUG.

Embodiment 21 is the method of embodiment 19 or embodiment 20, wherein the RNA is edited such that the methionine at position 148 of the encoded protein is replaced with valine.

Embodiment 22 is the method of any one of embodiments 19-21 , wherein the target RNA is edited in a cell.

Embodiment 23 is the method of embodiment 22, wherein the target RNA is edited in a liver cell.

Embodiment 24 is a method of restoring the function of the PNPLA3 protein in a subject characterised as having the PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding the oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR- mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

Embodiment 25 is use of ADAR-mediated RNA editing to convert a PNPLA3-148M protein variant to a PNPLA3-148V protein variant in subjects having liver disease or at risk of developing liver disease. Embodiment 26 is use of ADAR-mediated RNA editing to prevent or treat liver disease in subjects in need thereof, wherein the subjects are characterized as having a PNPLA3-148M protein variant and the ADAR-mediated RNA editing converts the PNPLA3-148M protein variant to a PNPLA3-148V protein variant.

Embodiment 27 is the method of any one of embodiments 1 -12 or 15-24 wherein the oligonucleotide is conjugated to N-acetylgalactosamine (GalNAc).

Embodiment 28 is the method of embodiment 27, wherein the oligonucleotide is conjugated to a triantennary N-acetylgalactosamine (GalNAc) ligand.

Embodiment 29 is a method of attenuating loss of lipase activity in a subject characterized as having a PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

Embodiment 30 is a method of reducing an amount of lipid accumulation in a subject characterized as having a PNPLA3 protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR-mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

Embodiment 31 is the method of embodiment 29 or embodiment 30, wherein the subject is a human subject.

Embodiment 32 is the method of any one of embodiments 29-31 , wherein prior to administration of the oligonucleotide or the DNA molecule, the method comprises a step of detecting a PNPLA3 rs738409 allele in a sample obtained from the subject.

Embodiment 33 is the method of any one of embodiments 29-32, wherein the subject is homozygous, or heterozygous, for the PNPLA3 rs738409 allele. Embodiment 34 is the method of any one of embodiments 29-33, wherein the polynucleotide encoding the PNPLA3-148M protein variant is a messenger ribonucleic acid (mRNA) molecule.

Embodiment 35 is the method of any one of embodiments 29-34, wherein the codon is converted by ADAR-mediated editing from AUG to IUG.

Embodiment 36 is the method of any one of embodiments 29-35, wherein the oligonucleotide is a guide RNA.

Embodiment 37 is the method of any one of embodiments 29-36, wherein the oligonucleotide is between 30 and 100 nucleotides in length.

Embodiment 38 is the method of any one of embodiments 29-37, wherein the oligonucleotide comprises or consists of an antisense nucleotide sequence at least partially complementary to the region of the polynucleotide encoding the PNPLA3-148M variant spanning position 148.

Embodiment 39 is the method of embodiment 38, wherein the antisense nucleotide sequence is 20-80 nucleotides in length.

Embodiment 40 is the method of any one of embodiments 29-39, wherein the oligonucleotide comprises one or more ADAR-recruiting domains.

Embodiment 41 is the method of any one of embodiments 29-40, wherein the oligonucleotide comprises one or more domains to improve stability.

Embodiment 42 is the method of any one of embodiments 29-41 , wherein the DNA molecule is comprised in a viral vector, optionally an adeno-associated viral vector.

Embodiment 43 is the method of any one of embodiments 29-42, wherein the DNA molecule is administered via lipid nanoparticles.

Embodiment 44 is the method of any one of embodiments 29-43, wherein the polynucleotide encoding the PNPLA3-148M protein variant only undergoes ADAR-mediated editing in the cells of the liver.

Embodiment 45 is the method of any one of embodiments 29-44, wherein the lipase activity of PNPLA3 in the liver of the subject is increased relative to an initial lipase activity level and/or restored to wild-type levels. EXAMPLES

The invention will be further understood with reference to the following non-limiting examples.

Example 1 : Lipid accumulation assay in cells overexpressinq PNPLA3 variants PNPLA3- 148V and PNPLA3-148M or PNPLA3-148I (wild-tvpe).

Lipid accumulation was measured in cells overexpressing PNPLA3 variants PNPLA3-148V and PNPLA3-148M or PNPLA3-148I (wild-type). The experiment was repeated in three different cellular models: (i) HuH-7 PNPLA3 knock-out (KO) clonal cells, (ii) primary human hepatocytes (PHH), and (iii) human induced pluripotent (iPS) cell-derived hepatocytes.

Lipid Accumulation Assay in HuH-7 PNPLA3 KO Cells

HuH-7 PNPLA3 KO cells were prepared and transduced with a BacMam viral vector carrying a PNPLA3 variant (either 1481, 148M, or 148V) for 24 hours, as described below. Cells were then either treated with free fatty acid (oleic acid or palmitic acid), or incubated with no fatty acid treatment for another 24 hours to induce the disease phenotype of NASH, followed by staining with Nile Red. Lipid content in the cells was then determined by measuring the Nile red intensity and normalized to nuclei count from Hoechst staining. The p values were calculated with TTEST. The statistical significance was also confirmed by an independent statistical analysis accounting for potential sources of variability (data not shown).

As can be seen from the “Lipid Intensity/Nuclei” reported in FIGs. 7A-7C, HuH-7 PNPLA3 KO cells overexpressing the PNPLA3-148M variant accumulated significantly more lipid than cells overexpressing the wild-type protein (PNPLA3-148I). In contrast, the levels of lipid seen in cells overexpressing the PNPLA3-148V variant were significantly lower than in cells overexpressing PNPLA3-148M and were similar to the levels seen in cells overexpressing wild-type PNPLA3- 1481.

Lipid Accumulation Assay in Primary Human Hepatocyte (PHH) Cells

Lipid accumulation was also measured in PHH cells overexpressing PNPLA3 variants. PHH cells derived from two donors (Hu8356 or HUM4167), respectively, that were heterozygous for PNPLA3 (148M/148I) were plated and transduced with a PNPLA3 variant BacMam (148V, 148M, or S47A 1481), or wild-type (1481), as described below. The PNPLA3-S47A 1481 is a catalytically inactive variant. Two days after BacMam transduction, lipid content in the cells was determined by staining cells with Lipidtox Green, measuring the Lipidtox Green intensity, and was normalized with nuclei count from Hoechst staining. The results are shown in FIGs. 8A-8B which demonstrate that PHH cells overexpressing PNPLA3-148V cells had decreased lipid accumulation as compared to cells overexpressing PNPLA3-148M or the catalytically inactive variant PNPLA3-S47A 1481. These data also indicate that PNPLA3-148M is a loss-of-function (LoF) mutation in the lipid accumulation assay, linking the lipase activity of PNPLA3 with lipid content in PHHs.

Lipid Accumulation Assay in iPS-Cell Derived Hepatocytes

Lipid accumulation was also measured iPS-cell derived hepatocytes, specifically FCDI iCell hepatocytes transduced with PNPLA3 variant BacMam (148V, 148M, or S47A 1481) or wild-type (1481), as described below. Two days after BacMam transduction, lipid content in the cells was determined by staining cells with Lipidtox Green, measuring the Lipidtox Green intensity, and was normalized with cell count from Hoechst staining. The results are shown in FIG. 9 and demonstrate that cells overexpressing PNPLA3-148V had significantly lower lipid accumulation that cells overexpressing PNPLA3-148M, whereas cells overexpressing PNPLA3-S47A 1481 (catalytically inactive variant) showed the highest amount of lipid accumulation. These data also indicated that PNPLA3-148M is a loss-of-function (LoF) mutation in the lipid accumulation assay, linking the lipase activity of PNPLA3 with lipid content in iPS-cell derived hepatocytes.

Conclusion

Taken together, these results demonstrate that overexpression of the PNPLA3 variants PNPLA3- 148V and PNPLA3-148I (wild-type) significantly lowers lipid accumulation in hepatic cells used compared to overexpression of PNPLA3-148M. This suggests that administration, or use, of oligonucleotides capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant to PNPLA3-148V (functional variant) could treat patients with NASH or other liver disease carrying PNPLA3-148M/M homozygous and/or PNPLA3-148M/I heterozygous alleles.

Materials and Methods

Generation of BacMam viral stocks for PNPLA3 variants: BacMam viral transfer vectors carrying PNPLA3 variants (with 1481, 148M, 148V or S47A 1481) were generated in the BacMam transfer vector pHTBVI .1 . BacMam P1 viral stocks were generated as previously described (Fornwald et al. Methods Mol. Biol., 2016, 1350: 95-116). BacMam P2 viral stocks were amplified in Super9 cells grown in SF900II-SFM + 5% HI FBS, and titered as previously described (Fornwald et al. 2016). All BacMam virus solutions were prepared to the same virus particle concentration to ensure the transduction at the same volume under the same multiplicity of infection (MOI). PNPLA3 BacMam transduction and treatment in HuH-7 PNPLA3 KO cells: Cells of HuH-7 PNPLA3 KO clonal cells (generated via CRISPR genome editing in HuH-7 cells) were maintained in DMEM/F12 (ThermoFisher, Ref# 11039-021) supplemented with 10% FBS (ThermoFisher, Ref#16000-044) at 37°C with 5% CO2. On the day of transduction, dissociated cells were resuspended in culture media to desired cell density. Cell/BacMam solution was plated in 96 well cell culture imaging plates (Greiner Bio-One, Cat#655090). On the following day, cells were treated with or without free fatty acid at a specific concentration. Oleic acid (Sigma, Cat# 01008) was prepared in DMSO at stock of 400 mM and stored at -20°C. On the day of treatment, the stock solution was mixed with cell culture grade BSA (Sigma, Cat# A9576) at a molar ratio of 4:1 (fatty acid: BSA), then diluted in culture media to 2 x of final concentration. An equal volume of prepared fatty acid media was added to the cell plate to reach the final desired concentration and incubated with cells for 24 hours. Palmitic acid (Cayman Chemicals, Cat#29558) was pre-mixed with BSA at a molar ratio of 6:1 and then diluted to 2 x of final concentration in culture media. An equal volume of the treatment was added to the cell plate to reach the final concentration and incubated with cells for 24 hours.

PNPLA3 BacMam transduction in primary human hepatocytes (PHHs): Frozen PHHs were thawed at 37°C and transferred to Hepatocyte Thaw medium (ThermoFisher, Cat# CM7500) and spun down at 100g for 10 minutes. Cells were resuspended in Hepatocyte plating medium, William’s E medium (ThermoFisher, Cat#A1217601) supplemented with plating medium supplements (ThermoFisher, Cat# CM3000). Cells were gently mixed and the cell density and cell viability was determined with a INCYTO C-chip Hemocytometer (Cat# DHC-N01). Cells were diluted in Hepatocyte medium to 0.4x10⁶ cells/mL. Diluted PHHs were plated at 100 pL/well (40 K cells/well) in 96 well collagen coated plates (Rewity, Cat# 6055700) which were pre-incubated with 100 pL/well of plating medium for 30 minutes at 37°C. Cell plates were incubated in a 37°C with 5% CO2 incubator for around 6 hours. After 6 hours incubation for PHH cell attachment, BacMam virus solutions were diluted to multiplicy of infection (MO 00) in Hepatocyte maintaining medium, William’s E Medium supplemented with Plating Medium Supplement (ThermoFisher CM4000). The plating medium was replaced with the prepared BacMam solutions at 100 pL/well. Cells were incubated for two days before fixation and staining.

PNPLA3 BacMam transduction in FCDI ICell Hepatocytes (IHep): FCDI iCell hepatocyte 2.0, short for FCDI iHep, were purchased from FUJIFILM Cellular Dynamics (FCDI, R1 104) and cultured following the manufacturer’s guideline. FCDI iHep cells are induced pluripotent stem (iPS)-cell derived hepatocyte cells. Briefly, FCDI iHep were thawed into RPMI 1640-based plating medium (ThermoFisher, 11875) containing B-27 Supplement (ThermoFisher, 17504), recombinant human oncostatin M at 10 pg/ml (R&D Systems, 295-OM), dexamethasone at 5 mM (ThermoFisher, ICN19456125), gentamicin at 50 mg/mL (ThermoFisher, 15750), and iCell Hepatocytes 2.0 Medium Supplement (FCDI, M1024). FCDI iHep were plated at 300,000 viable cells/cm² into collagen-coated plates (Perkin Elmer LLC, 50-209-9828) and cultured at 37°C, 5% CO2 in plating medium for 5 days with 100% medium change every day. At day 5, cells were transduced with various BacMams expressing PNPLA3-148I, -148M, -148V, and S47A 1481, respectively, at MOI=20.

Cell fixation, staining, and imaging analysis in HuH-7 PNPLA3 KO cells and PHHs: Two days after BacMam transduction, cells (with and/or without free fatty acid treatment) were washed two times with PBS and fixed with 3.75% Formaldehyde (Electron Microscopy Sciences, Cat#15682) in PBS for 15 min at room temperature. After washing two times, fixed cells were incubated with 3 pg/mL Nile red (Sigma, Cat#19123) for HuH-7 PNPLA3 KO cells, or with LipidTox Green (ThermoFisher, Cat#H34475) at 2000x dilution for PHHs, together with 3 pM Hoechst (ThermoFisher, Cat#62249) diluted in PBS for 30 min at room temperature. Stained cells were washed twice and imaged with Opera Phenix Plus high content imager at 40x. Cell nuclei were identified and counted from Hoechst channel (Ex375 nm Em435-480 nm). Both Nile red and LipidTox Green signals were scanned at Ex488 nm Em 500-550 nm) for lipid content. The measured lipid content was normalized per nuclei number. Results were graphed with GraphPad Prism.

Cell fixation, staining, and imaging analysis in FCDI iCell Hepatocytes-. Two days after BacMam transduction, cells were washed with DPBS twice and fixed with 4% paraformaldehyde (Fisher scientific, 50-980-495) in DPBS for 20 min at room temperature. After three washes with DPBS, cells were stained with solution containing LipidTox Green neutral lipid stain and Hoechst 33342 in DPBS for 20 min at room temperature in the dark. Cells were imaged using Opera Phenix Plus high content imager and images were analyzed with SignalsImageArtist. Results were graphed with GraphPad Prism.

Example 2: Lipase Activity of Purified PNPLA3 Variants

The lipase activity of four purified recombinant fusion proteins of full-length and truncated (1-276) PNPLA3 variants (1481, 148M, 148V, 1481 S47A) was measured using both a triglyceride substrate and a monoglyceride substrate. The recombinant fusion proteins were expressed and purified as described below, and then tested in two different lipase assays using either a triglyceride substrate or a monoglyceride substrate, as described below.

The results of the lipase activity assay for the fusion protein with the full-length PNPLA3 variants are shown in FIGs. 10A and 10B and the results of the lipase activity assay for the fusion protein with truncated PNPLA3 variants are shown in FIGs. 11 A and 11 B. For the full length PNPLA3 proteins, the derived values for kcat for 1481 (wild-type) and 148V (variant) were within 1 .2-fold in the TG assay and within 1 .4-fold in the MG assay (FIGs. 10A and 10B). Notably, the kcat for 148V is 4.5-fold faster than 148M in the TG assay and 3.1 -fold faster than 148M in the MG assay. The small residual activity observed with S47A 1481 is likely due to a small amount of a contaminating lipase that can hydrolyze the MG substrate, but not the TG substrate. For the truncated PNPLA3 proteins, the derived values for kcat for 1481 (wild-type) and 148V (variant) were within 1 .1 -fold in both assays. Notably, the kcat for 148V is 4.8-fold faster than 148M in the TG assay. In the MG assay, 148V is 1 .4-fold faster than 148M (FIGs.. 11 A and 11 B). The solubility of the MUB substrate places an upper limit on the highest concentration that can be tested in that assay format, and thus the substrate MG K_m for 148M is poorly defined.

Taken together, the results demonstrate that the PNPLA3-148V variant has significantly improved enzymatic lipase activity as compared to the 148M variant and catalytically inactive S47A 1481 variant, with activity similar to that of WT (1481).

Materials and Methods

Generation ofE. coli expression constructs forPNPLA3 variants: Human PNPLA3 full length variants were generated using a synthetic approach. PNPLA3 DNA sequence was codon optimized for E. coli expression. The Trigger Factor chaperone (TF) was used as a soluble tag. Each gene was synthesized and cloned into pColdTF (TaKaRa, US) between Ndel and Hind3 restriction enzyme sites to generate recombinant plasmid.

Protein expression: BL21 Competent E. coli (New England BioLabs) were transformed with the recombinant plasmid according to the standard transformation method. A single colony was inoculated into 50 mL of LB medium (Gibco) containing 100 pg/mL carbenicillin (Teknova) and incubated with shaking (220 rpm) at 30°C overnight. An aliquot of the overnight culture was inoculated into 1 L of Terrific Broth Complete (Teknova) containing 100 pg/mL carbenicillin and 1 :20,000 dilution of Antifoam 204 (Sigma) to get the starting OD600 = 0.10, incubated with shaking (260 rpm) at 30°C until the OD600 reached 0.60 - 0.80, and then the culture was cooled down to 16°C. Then the culture was induced with 0.5 mM IPTG and incubated with shaking (260 rpm) at 16°C for 18-20 hours. The cells were harvested by centrifugation (4,000 rpm at 4°C for 10 min).

Purification of full length His-TF-PNPLA3-K434E fusion proteins-. Four (4) fusion proteins were purified by NiNTA agarose resin (Cube Biotech) and Superose6 increase 10/300G size exclusion column (Cytiva life science). In detail, each cell paste from 0.5 L of E. coli culture was resuspended in lysis buffer (50 mM Tris, pH8.0, 300 mM NaCI, 20 mM Imidazole, 0.1 mM TCEP with Protein inhibitor tablets from Roche) and lysed by sonication. The lysate was centrifuged for 30min at 30,000 x g at 4 °C to eliminate cell debris. The supernatant was then mixed with NiNTA agarose resin overnight at 4°C. The resin was washed with lysis buffer containing 35mM imidazole. Protein was eluted from the Ni resin by 300 mM imidazole elution buffer and further purified by Superpose 6 size exclusion column (buffer 25 mM Tris, pH 7.5, 150 mM NaCI, 1 mM EDTA, 0.1 mM TCEP). Two major peak fractions were pooled and the purified protein from the pooled fractions was used in the lipase activity assays.

Purification of His-TF-PNPLA3(1-276) fusion proteins'. Four (4) PNPLA3(1-276) truncation fusion proteins were purified the same as for the full-length proteins. There was only one single elution peak from the sizing column. Instead of pooling all fractions together, three elution fractions were selected from the right shoulder to minimize collection of aggregated protein. The purified protein from the pooled fractions was used in the lipase activity assays.

All purified PNPLA3 fusion protein samples were analyzed by SDS-PAGE gel to confirm the expected size and purity. Protein concentration was determined by Bradford assay. LC-MS detected that all the truncated proteins had the expected mass. LC-MS-MS peptide mapping of the protein bands separated by SDS-PAGE gel confirmed the correct amino acid sequence for the WT and mutant proteins.

Assay 1-Lipase activity assay with a quenched fluorescent triglyceride (TG) substrate'. The triglyceride (TG) lipase activity was performed at room temperature with a BMG Labtech PHERAStar FS(X) plate reader in kinetic mode, by monitoring the hydrolysis of the quenched fluorescent triglyceride substrate 1-oleoyl-2-(6-(2,4-dinitrophenyl)amino)hexanoyl-3-[11- (dipyrrometheneboron)undecanoyl]-sn-glycerol (18:1-6:0 DNP-C11 TopFluorR TG; Avanti Polar Lipids) through fluorescence excitation at 485 nm and emission at 520 nm by a modification of previously described methods (Basu et al. J. Lipid Res. 2011 , 52(4): 826-32; Haller et all J. Lipid Res. 2017, 58(6): 1166-1173). The total reaction volume in 384-well Greiner black low volume microtiter plates was 10 pL of 10 mM phosphate-buffered saline (PBS) pH 7.4 containing 0.005% Zwittergent 3-14 (EMD Millipore), 0.1 mM tris(2-carboxyethyl)phosphine hydrochloride (TCEP; Thermo), 0.15% (w/v) fatty acid free bovine serum albumin (BSA; Millipore-Sigma), 0.1 to 150 pM 18:1-6:0 DNP-C1 1 TopFluorR TG, in 2-fold increments and 300 nM PNPLA3 FL or 100 nM PNPLA3 (1-276) constructs (n=4). The top concentration of the TG substrate was prepared in assay buffer from a DMSO stock in a glass vial at 2x final concentration by vortexing for 20 sec, sonicating for 10 min in a Fisher ultrasonic bath then additional vortexing for 20 sec. Subsequent dilutions were prepared in glass with vortexing for 20 sec between dilutions in assay buffer containing 0.75% dimethylsulfoxide (DMSO; Millipore-Sigma). Prior to adding 5 pL of the substrate to the assay plate, the substrate dilutions were vortexed for 20 sec, sonicated for 10 minutes, and vortexed for an additional 20 sec. Reactions were initiated with the addition of 5 pL of 2x final concentrations of PNPLA3 constructs. A standard curve of 1 -oleoyl-3-[1 1- (dipyrrometheneboron difluoride)undecanoyl]-rac-glycerol (18:1 -C11 TopFluorR TG; Avanti Polar Lipids) concentrations (0.01 to 10 pM in 2-fold increments) versus fluorescence was prepared to allow the conversion of changes in fluorescence to pM product formed. The initial reaction rate for PNPLA3 activity was determined by calculating the slope of the linear progress curves using the MARS data analysis package, and then the calibration curve was used to convert RFU/min to pM/min.

Assay 2-Lipase Assay with a fluorescent monoglyceride (MG) substrate: The monoglyceride (MG) lipase activity was performed at room temperature with a BMG Labtech PHERAStar FS plate reader in kinetic mode, by monitoring the hydrolysis of the fluorescent substrate 4- methylumbelliferone heptanoate (heptanoate-MUB; Millipore-Sigma) through fluorescence excitation at 380 nm and emission at 470 nm by a modification of previously described methods for measuring thioesterase activity (Richardson and Smith, Mol. Cancer Ther. 2007, 6(7), 2120- 2126; Hardwicke et al. Nat. Chem. Biol. 2014, 10, 774-779). The total reaction volume in a 384- well Greiner black low volume microtiter plates was 10 pL of 10 mM PBS pH 7.4 containing 0.03% 3-[(3-cholamidopropyl)dimethylammonio]-1 -propanesulfonate hydrate (CHAPS; Millipore- Sigma), 0.1 mM TCEP, 0.26 to 10 pM heptanoate-MUB in 1.5-fold increments and 100 nM PNPLA3 FL (n=2) or 25 nM PNPLA3 (1-276) (n=4) constructs. Substrate dilutions from DMSO stocks were prepared in assay buffer at 2x final concentration in a polypropylene plate followed by pipetting 5 pL into an assay plate. Reactions were initiated by adding 5 pL of 2x final concentration of PNPLA3 constructs. A standard curve of 4-methylumbelliferone (4-MU; Millipore-Sigma) concentrations (0.02 to 5 pM in 2-fold increments) vs fluorescence was prepared to allow the conversion of changes in fluorescence to pM 4-MU produced. PNPLA3 activity was calculated from the slope of the linear region of the progress curves using the MARS data analysis software package. The no-enzyme control background rates were subtracted and expressed as initial rates in pM/min after converting RFU/min with the 4-MU standard curve.

Data analysis for Assay 1 and Assay 2 Values for the Vmax and K_m parameters and their standard errors were obtained fitting Equation (1) to the average initial rate data, using the nonlinear regression function of GraFit (Erithacus Software). v > ^vmax*[s] Equation (1)

K_m + [S]

In Equation (1), v is the initial velocity, Vmax is the maximum velocity, S is the varied substrate concentration, and K_m is the Michaelis constant. To compare the different constructs, the parameters were normalized to their protein concentrations, calculating kcat assuming 100% PNPLA3 and that the corresponding fractions for each construct had similar purity based on the SDS-PAGE analysis.

Claims

1 . A method of preventing or treating liver disease in a subject characterised as having a PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding an oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR- mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

2. The method of claim 1 , wherein the subject is a human subject.

3. The method of claim 1 or claim 2, wherein prior to administration of the oligonucleotide or the DNA molecule, the method comprises a step of detecting a PNPLA3 rs738409 allele in a sample obtained from the subject.

4. The method of any one of claims 1 -3, wherein the subject is homozygous, or heterozygous, for the PNPLA3 rs738409 allele.

5. The method of any one of claims 1-4, wherein the polynucleotide encoding the PNPLA3-148M protein variant is a messenger ribonucleic acid (mRNA) molecule.

6. The method of any one of claims 1-5, wherein the codon is converted by ADAR-mediated editing from AUG to IUG.

7. The method of any one of claims 1-6, wherein the oligonucleotide is a guide RNA.

8. The method of any one of claims 1-7, wherein the oligonucleotide is between 30 and 100 nucleotides in length.

9. The method of any one of claims 1-8, wherein the oligonucleotide comprises or consists of an antisense nucleotide sequence at least partially complementary to the region of the polynucleotide encoding the PNPLA3-148M variant spanning position 148.

10. The method of claim 9, wherein the antisense nucleotide sequence is 20-80 nucleotides in length.

11 . The method of any one of claims 1 -10, wherein the oligonucleotide comprises one or more ADAR-recruiting domains.

12. The method of any one of claims 1-11 , wherein the oligonucleotide comprises one or more domains to improve stability.

13. The method of any one of claims 1-12, wherein the DNA molecule is comprised in a viral vector, optionally an adeno-associated viral vector.

14. The method of any one of claims 1 -13, wherein the DNA molecule is administered via lipid nanoparticles

15. The method of any one of claims 1 -14, wherein the polynucleotide encoding the PNPLA3- 148M protein variant only undergoes ADAR-mediated editing in the cells of the liver.

16. The method of any one of claims 1-15, wherein the lipase activity of PNPLA3 in the liver of the subject is increased relative to an initial lipase activity level and/or restored to wild-type levels.

17. The method of any one of claims 1-16, wherein the liver disease is selected from nonalcoholic fatty liver disease (NAFLD), alcoholic-related liver disease (ArLD), liver fibrosis and cirrhosis.

18. The method of claim 17, wherein the NAFLD is nonalcoholic steatohepatitis (NASH) or metabolic dysfunction associated steatohepatitis (MASH).

19. A method of editing a target ribonucleic acid (RNA) molecule encoding a PNPLA3-148M protein variant, the method comprising contacting the target RNA with an oligonucleotide capable of effecting ADAR-mediated adenosine to inosine alteration of the codon encoding the methionine at position 148, thereby editing the target RNA.

20. The method of claim 19, wherein the codon encoding the methionine at position 148 is converted from AUG to IUG.

21 . The method of claim 19 or claim 20, wherein the RNA is edited such that the methionine at position 148 of the encoded protein is replaced with valine.

22. A method of restoring the function of the PNPLA3 protein in a subject characterised as having the PNPLA3-148M protein variant, the method comprising: administering to the subject an oligonucleotide or a DNA molecule encoding the oligonucleotide, wherein the oligonucleotide is capable of effecting ADAR-mediated editing of a polynucleotide encoding the PNPLA3-148M protein variant, and wherein the polynucleotide encoding the PNPLA3-148M protein variant undergoes ADAR- mediated editing at a codon encoding the methionine at position 148 such that the polynucleotide encodes a PNPLA3-148V protein variant.

23. Use of ADAR-mediated RNA editing to convert a PNPLA3-148M protein variant to a PNPLA3-148V protein variant in subjects having liver disease or at risk of developing liver disease.

24. Use of ADAR-mediated RNA editing to prevent or treat liver disease in subjects in need thereof, wherein the subjects are characterized as having a PNPLA3-148M protein variant and the ADAR-mediated RNA editing converts the PNPLA3-148M protein variant to a PNPLA3-148V protein variant.

25. The method of any one of claims 1-11 wherein the oligonucleotide is conjugated to N- acetylgalactosamine.