Targeted Epigenetic Modulation Outperforms Nuclease- and Deaminase-Based Editing for Durable Pcsk9 Silencing in a Clinically Relevant Delivery System

This study demonstrates that targeted epigenetic modulation, delivered via Arcturus LUNAR lipid nanoparticles, achieves superior and durable PCSK9 silencing in mice compared to cytosine base editing, CRISPR-Cas9, and GalNAc-siRNA, establishing it as a promising strategy for long-lasting hepatic therapy.

The Gist

Imagine your liver is a bustling factory that manages your body's cholesterol. One of the factory's most important managers is a protein called PCSK9. Normally, this manager is helpful, but when there's too much of it, it starts destroying the factory's "trash collectors" (receptors) that clean up bad cholesterol from your blood. Too much PCSK9 means bad cholesterol builds up, leading to heart disease.

For years, doctors have tried to stop this manager from causing trouble using different tools:

1. The "Stop Sign" (Antibodies/SiRNA): These are like temporary roadblocks. They work great for a while, but they wear off quickly, so you have to keep taking them every few weeks or months.
2. The "Scissors" (CRISPR-Cas9): This is a gene editor that cuts the DNA blueprint of the PCSK9 manager to break the factory's instructions permanently. It's powerful, but cutting DNA is risky; it's like using a chainsaw to fix a watch. You might accidentally cut the wrong thing or cause a messy repair job (mutations).
3. The "Pencil" (Base Editing): This is a more precise tool that tries to change just one letter in the DNA instruction manual to stop the manager. It's safer than the chainsaw, but it's still trying to rewrite the permanent blueprint, which can be tricky and sometimes doesn't stick around long enough.

The New Discovery: The "Silence Button" (Epigenetic Editing)

In this new study, scientists from Arcturus Therapeutics tried a completely different approach. Instead of cutting the DNA or rewriting the letters, they invented a "Silence Button."

Think of the DNA as a book of instructions.

• Cutting (CRISPR) tears out the page.
• Rewriting (Base Editing) changes the words on the page.
• Silencing (Epigenetic Editing) puts a heavy, sticky note over the page that says, "Do not read this." The words are still there, but the factory workers (the cell's machinery) can't see them, so they stop making the PCSK9 manager.

How They Tested It

The researchers put this "Silence Button" into a special delivery truck called LUNAR® (a type of lipid nanoparticle). This truck is designed to drive straight into the liver cells and drop off the instruction manual for the Silence Button.

They tested three teams in mice:

1. Team Scissors (CRISPR): Cut the DNA.
2. Team Pencil (Base Editor): Tried to change a letter.
3. Team Silence (Epigenetic Editor): Put the sticky note over the instructions.

The Results

• Team Scissors and Team Pencil did a decent job at first, but their effects started to fade after a few weeks. It's like the factory workers eventually found a way to read around the torn page or the pencil marks, and the PCSK9 manager started showing up again.
• Team Silence was the clear winner. By putting that "Do Not Read" note on the DNA, they completely shut down the PCSK9 manager. Even better, this silence lasted for the entire 30-day study (and likely much longer). The factory workers couldn't get the instructions, so they stopped making the problem protein entirely.

Why This Matters

The study suggests that the "Silence Button" is the safest and most effective way to treat high cholesterol with a single shot.

• Safety: Because it doesn't cut the DNA, there's a much lower risk of accidentally breaking the factory's other important blueprints.
• Durability: It works longer than the current "roadblock" drugs (which you have to take often) and seems more reliable than the "cutting" tools.
• Simplicity: It uses a single injection of mRNA (the instruction manual) rather than complex, multi-part systems.

The Bottom Line

This research shows that sometimes, the best way to fix a problem isn't to destroy the instructions or rewrite them, but simply to make them unreadable. This "Epigenetic Silencing" approach, delivered by the LUNAR® truck, could lead to a "one-and-done" cure for high cholesterol, where a single injection keeps your heart healthy for years without the need for constant medication or the risks of permanent DNA cutting.

Technical Summary

1. Problem Statement

Inhibition of Proprotein Convertase Subtilisin/Kexin Type 9 (PCSK9) is a validated strategy for lowering Low-Density Lipoprotein (LDL) cholesterol and reducing cardiovascular risk. While existing therapies exist, they have limitations:

• Monoclonal antibodies require frequent dosing (every 2–4 weeks).
• GalNAc-siRNA (e.g., Inclisiran) offers longer duration but still requires semi-annual dosing and does not constitute a true "one-and-done" cure.
• Genome Editing (CRISPR-Cas9 and Base Editing): While promising for permanent correction, these modalities face significant safety hurdles. Nuclease-based editing introduces double-strand breaks (DSBs), risking chromosomal rearrangements and indels. Base editing avoids DSBs but can still cause off-target mutations and bystander edits. Furthermore, achieving high editing efficiency in a single administration to silence a gene permanently in a large population of hepatocytes remains challenging.

The study aims to compare the efficacy, durability, and safety of epigenetic editing against CRISPR-Cas9, Cytosine Base Editing (CBE), and GalNAc-siRNA for Pcsk9 suppression, utilizing a clinically relevant, non-viral mRNA delivery system.

2. Methodology

A. Modalities Developed

The researchers engineered three distinct gene-editing modalities, all delivered as mRNA:

1. Cytosine Base Editor (CBE): Designed to introduce a premature stop codon in the Pcsk9 open reading frame (ORF).
   • Components: High-fidelity SpCas9 nickase (D10A) fused to Petromyzon marinus cytidine deaminase (PmCDA1) and a uracil-DNA glycosylase inhibitor (UGI).
   • Target: Exon 1 of mouse Pcsk9, specifically targeting Tryptophan (TGG) codons to convert them to Stop codons (TAA/TAG/TGA) via C-to-T conversion on the non-complementary strand.
2. CRISPR-Cas9 Gene Editor: A standard nuclease-based editor using SpCas9-HF (high-fidelity) to induce DSBs and disrupt the gene via indels.
3. Epigenetic Editor (Epi-Editor): A transcriptional repressor that does not alter the DNA sequence.
   • Components: A Zinc Finger Protein (ZFP) DNA-binding domain targeting the Pcsk9 5'UTR CpG island, fused to a dual-effector domain: DNMT3A (DNA methyltransferase) and KRAB (transcriptional repressor).
   • Strategy: The study tested single domains (DNMT only, KRAB only) and the dual-domain fusion to determine the optimal configuration for stable silencing.

B. Delivery System

• LUNAR® Lipid Nanoparticles (LNPs): All mRNA constructs were formulated using Arcturus Therapeutics' proprietary ionizable lipid LNP system. This system is designed for hepatocyte-specific delivery, rapid manufacturing, and transient expression (minimizing immune response).
• Controls: Included GalNAc-conjugated siRNA (Inclisiran-like) and eGFP mRNA controls.

C. Experimental Models

• In Vitro: Hepa1-6 mouse hepatocyte cells were used for dose-response studies, durability assays (up to 21 days), and sequencing analysis.
• In Vivo: Wild-type C57BL/6N female mice received a single intravenous injection of 1 mg/kg. Blood was collected over 30 days to measure plasma PCSK9, LDL-C, and total cholesterol. Liver toxicity (AST/ALT) was monitored.

3. Key Contributions

1. Optimization of Epigenetic Silencing: The study demonstrated that dual-effector epigenetic editors (DNMT-KRAB) are superior to single-effector domains. Crucially, it revealed that the temporal coordination of these domains is vital; sequential delivery showed that KRAB followed by DNMT (KRAB → DNMT) yielded the most durable silencing, suggesting histone modification primes the locus for stable DNA methylation.
2. Direct Modality Comparison: This is one of the first studies to directly benchmark epigenetic editing against advanced base editing and high-fidelity nuclease editing within the same delivery framework (LNP-mRNA) and target (Pcsk9).
3. Validation of LUNAR® for Epigenetics: The study proved that the LUNAR® delivery system can effectively deliver complex epigenetic editors to hepatocytes, achieving near-complete silencing with a favorable safety profile compared to previous LNP formulations used in similar studies.

4. Key Results

In Vitro Findings

• Epigenetic Superiority: The dual-domain (DNMT-KRAB) epi-editor achieved an IC₅₀ of 0.09 nM, significantly outperforming single-domain editors. It reduced Pcsk9 transcripts to 5.3% of control levels at maximal effect, whereas CRISPR-Cas9 and CBE only achieved 29.7% and 53.6% reduction, respectively.
• Durability: Only the dual-domain epi-editor maintained robust repression for 21 days. Single-domain editors and the gene editors showed transient effects or failed to sustain silencing as cells divided.
• Mechanism: The epi-editor showed a strong correlation between mRNA and protein suppression, whereas the gene editor showed a divergence (high protein suppression but less transcriptional suppression), indicating that indels primarily blocked translation rather than transcription.

In Vivo Findings (Mice)

• PCSK9 Suppression:
  • Epi-Editor: Achieved ~100% reduction in plasma PCSK9 at Day 3 and maintained near-complete suppression throughout the 30-day study.
  • Gene Editor: Achieved ~81% reduction at Day 3 but declined rapidly, falling to ~50% by Day 30.
  • Base Editor: Achieved 48% reduction at Day 3, dropping to minimal residual activity (11%) by Day 30.
  • GalNAc-siRNA: Achieved ~78% reduction at Day 7, declining to ~60% by Day 30.
• Lipid Profile: While PCSK9 was suppressed, LDL-C and total cholesterol reductions were modest and often not statistically significant in wild-type mice (due to low baseline LDL and compensatory mechanisms). However, the epi-editor group consistently showed the greatest trend toward cholesterol reduction.
• Safety: All LNP formulations were well-tolerated. AST and ALT levels remained comparable to PBS controls, indicating no overt hepatotoxicity. The LUNAR® formulation showed a significantly better safety profile than LNP formulations used in previous epigenetic studies (which caused 8-10x enzyme elevations).

5. Significance and Conclusion

• Superior Durability: The study establishes epigenetic editing as a more durable therapeutic strategy than nuclease or base editing for Pcsk9 silencing in this context. The ability to maintain near-complete silencing for 30 days (and likely longer given the epigenetic nature) with a single dose supports the "one-and-done" therapeutic goal.
• Safety Profile: By avoiding double-strand breaks and permanent genomic mutations, epigenetic editing mitigates the risks of chromosomal rearrangements and off-target mutagenesis associated with CRISPR-Cas9 and base editors.
• Delivery Innovation: The successful use of LUNAR® LNPs demonstrates that non-viral delivery of complex epigenetic machinery is feasible and safe, offering a scalable alternative to viral vectors (AAV).
• Therapeutic Implication: This work provides a strong proof-of-concept for using mRNA-encoded, dual-domain epigenetic editors as a next-generation therapy for hypercholesterolemia and other diseases requiring long-term gene silencing without permanent DNA alteration.

Limitations Noted: The study utilized wild-type mice, which have low baseline LDL-C, potentially masking the full therapeutic impact on cholesterol levels. Future studies in hyperlipidemic models or non-human primates are recommended to fully validate the clinical lipid-lowering potential. Additionally, the base editor used was not the most optimized version available, suggesting room for improvement in that comparator arm.