Rational design of next-generation LNPs based on piperazine or cyclohexane ionizable lipids for effective protein replacement in rare metabolic disorders = 희귀대사질환의 효율적 단백질대체요법을 위한 피페라진 및 사이클로헥산 이온화 지질기반 차세대 LNP의 합리적 설계

다국어 초록 (Multilingual Abstract)

Lipid nanoparticles (LNPs) have demonstrated significant utility in the clinical applications of messenger RNA (mRNA), as evidenced by the regulatory approval of mRNA/LNPs COVID-19 vaccines. The technological advancement has facilitated research into diverse therapeutic applications, encompassing not only prophylactic vaccines but also protein replacement therapy, genome editing, and cancer immunotherapy. However, many existing lipid formulations were originally designed for short-term vaccine applications and are consequently not optimized for long-term therapeutic use. This presents a significant challenge for mRNA-based chronic therapies: the diminishing efficacy and escalating toxicity often associated with repeated dosing of LNPs. To overcome these limitations, we developed a next-generation ionizable lipid library platform engineered to enhances therapeutic protein expression efficiency and mitigate toxicity through structure-activity relationship (SAR)-driven design. In this study, we designed piperazine- and cyclohexane-based ionizable lipid libraries and elucidated the SAR through extensive in vitro and in vivo efficacy evaluations. These comprehensive studies aimed not only to identify high-potency delivery systems but also to establish fundamental design principles for chronic mRNA therapeutics.
The first phase of this research focused on mapping the chemical space of ionizable lipids by synthesizing a comprehensive library with diverse head, link, and tail groups. Subsequent SAR analysis revealed that functional performance was significantly influenced by scaffold geometry, identifying distinct structural advantages. Within the piperazine library, we observed that lipids featuring hydroxylated headgroups and ester-linked-tails- exemplified by candidates such as PA12 and PC27 – facilitated high initial protein expression. This was attributed to enhanced mRNA interactions and optimized endosomal escape kinetics. Furthermore, these piperazine-based lipids exhibited a significantly improved biocompatibility profile. In direct comparisons, they demonstrated markedly lower hepatotoxicity and pro-inflammatory cytokine induction than established piperazine-headed benchmarks, such as C12-200, and CKK-E12, which have historically been constrained by narrow therapeutic windows. In contrast, the cyclohexane-based lipids, particularly CA09 and CA10, demonstrated superior performance in chronic, repeated-dose animal studies. Our mechanistic investigations suggested that this enhanced durability stemmed from the scaffold’s inherent rigidity and the hydrolytic stability of the amide linkers. These structural features collectively promoted consistent nanoparticle morphology and prevented premature dissociation of the mRNA payload. This improved structural stability translated into durable protein expression and high tolerability, directly fulfilling the requirements for chronic mRNA-based enzyme replacement therapies.
The second phase of the study validated the GCP LNPs in the context of Phenylketonuria (PKU), a monogenic liver disorder caused by deficient phynylalanine hydroxylase (PAH) activity. When encapsulated with lead cyclohexane-based lipid (CA10), the optimized unmodified PAH mRNA achieved profound metabolic correction in the Paheun2 mouse model. A single administration resulted in a rapid clearance of serum Phe, reaching a reduction of over 95% within 6 hours. Notably this therapeutic effect was superior to clinically validated MC3 LNP benchmark. This study established a proof-of-concept for the systemic delivery of PAH mRNA as a potent modality for metabolic restoration in PKU and underscored the synergy between the refined lipid scaffold and the tailored mRNA payload.
The third phase of this research addressed Succinic Semialdehyde Dehydrogenase (SSADH) deficiency, a rare neuro-metabolic disorder. The therapeutic rationale rested on the “metabolic sink” hypothesis: that restoration of hepatic enzyme function could modulate the toxic metabolites (GABA and GHB), thereby alleviating central nervous system (CNS) pathology without requiring direct blood-brain barrier crossing. In the SSADH-D mouse model, the nucleoside-modified mRNA encapsulated within the CA09 LNP demonstrated a highly durable pharmacodynamic profile. A single administration was sufficient to normalize systemic metabolite levels for at least 14 days. This sustained efficacy supported the feasibility of reduced dosing frequencies, which was a critical consideration for the long-term management of chronic enzyme deficiencies.
Beyond biological potency, the clinical success of an LNP platform depends on its stability and manufacturability. We conducted extensive formulation optimization to identify the C11 composition (molar ratio 25:10:50:2). This formulation was subjected to rigorous stability testing, demonstrating that it maintained physical integrity (particle size <100nm, PDI <0.2, and encapsulation efficiency >90%) and functional potency for at least six months when stored at –80°C.
In summary, this dissertation presented the GCP-LNP platform as a highly potent, safe, and versatile system for the delivery of therapeutic mRNA. By elucidating the SAR of piperazine and cyclohexane scaffolds, this work established a framework for designing ionizable lipids tailored for chronic applications rather than transient vaccination. The successful metabolic correction in both PKU and SSADH-D mouse models provided a compelling proof-of-concept for hepatic enzyme replacement therapy. While these results were promising in rodent models, future work will focus on validating these SAR trends in non-human primates (NHPs) to account for interspecies differences in protein corona and hepatocyte uptake kinetics. Additionally, long-term repeat-dosing studies will be required to fully define the safety profile over extended treatment duration. Ultimately, this research underscored the potential of rational LNP design to expand the reach of mRNA therapeutics from infectious diseases to rare genetic disorders, providing a viable therapeutic foundation for patients with limited therapeutic options.

국문 초록 (Abstract)

mRNA-LNP 기반 코로나19 백신이 FDA에 허가됨에 따라 지질 나노입자 (LNP)의 효능과 안전성이 임상에서 입증되었으며, 백신 뿐만 단백질 보충 요법, 유전자 편집, 암 면역 치료 등 다양한 질환의 치료제로의 응용 범위를 넓히고 있다. 하지만 기존 LNP 플랫폼은 주로 단기 투여용 백신에 최적화되어 있어, 장기적인 단백질 보충이 필요한 만성 질환 치료에는 한계가 있다. 특히 반복 투여 시 발생하는 효능 저하와 독성 증가는 mRNA 기반 치료제의 주요 과제로 남아 있으며, 그 과정에서 이온화 지질 (ionizable lipid)의 구조적 특성이 효능과 안전성을 결정짓는 핵심 요소로 작용한다.
본 연구에서는 이러한 한계를 극복하기 위해 차세대 이온화 지질 라이브러리 플랫폼을 구축하고, 구조-활성 관계 (SAR) 분석을 통해 단백질 발현 효율은 높이면서 독성은 완화된 최적의 지질 구조를 도출하고자 하였다. 이를 위해 피페라진 (Piperazine) 및 사이클로헥산 (Cyclohexane) 기반의 이온화 지질 라이브러리를 설계하였으며, 광범위한 in vitro 및 in vivo 효능 평가를 통해 지질 구조와 생물학적 활성 간의 상관관계를 규명하였다.
연구의 첫 번째 단계에서는 이온화 지질 구조 내 다양한 head-linker-tail 그룹을 조합하여 최적의 구조를 도출하고자 하였다. SAR 분석 결과, 지질의 골격 구조가 기능적 성능에 결정적인 영향을 미친다는 것을 확인하였다. 피페라진 기반 라이브러리 중 수산화된 head와 ester linker가 도입된 tail을 가진 이온화 지질 (예: PA12, PC27)은 초기 단백질 발현을 극대화하는데 효과적이었으며, 이는 향상된 mRNA 상호작용과 최적화된 엔도좀 탈출 성능에 기인한 것으로 예상된다. 특히 이들 지질은 동일한 피페라진 head를 가진 벤치마크 lipid인 C12-200 및 CKK-E12에 비해 현저히 낮은 간독성을 보이며 우수한 생체 적합성을 입증하였다. 반면, 사이클로헥산 기반 지질 (CA09, CA10)은 만성적인 반복 투여 모델에서 탁월한 효능을 나타냈다. 이러한 효능은 사이클로헥산 골격 특유의 강성 (rigidity)와 amide linker의 가수분해 안정성에서 비롯된 것으로 보이며, 이를 통해 나노입자의 형태적 안정성을 유지하고 mRNA의 조기 해리를 효과적으로 방지할 수 있었던 것으로 예상된다. 이러한 결과는 사이클로헥산 기반 LNP가 지속적인 단백질 발현과 낮은 누적 독성이 요구되는 만성 질환 치료에 매우 적합한 성능을 갖추고 있음을 시사한다.
연구의 두 번째 단계에서는 페닐케톤뇨증 (PKU) 모델을 통해 본 플랫폼 (GCP-LNP)의 치료적 유효성을 검증하였다. 사이클로헥산 기반의 선도 지질인 CA10기반 LNP에 최적화된 PAH mRNA를 봉입하여 Pahenu2 마우스에 투여한 결과, 단회 투여만으로도 6시간 이내에 혈청 페닐알라닌 수치가 95% 이상 감소하였다. 이는 기존의 임상적 벤치마크인 MC3 LNP 보다 우수한 수치였으며, 정교하게 설계된 이온화 지질 기반의 LNP가 선천성 대사 질환의 전신 치료에 얼마나 효과적인지를 보여주었다.
세번째 단계에서는 희귀 신경 대사 질환인 숙신알데히드 탈수소효소 (SSADH) 결핍증으로 연구 범위를 확장하였다. 본 연구는 간에서의 SSADH 효소 기능 회복이 전신 대사산물 (GABA, GHB)의 농도를 조절하여 뇌혈관 장벽 (BBB)을 통과하지 않고도 중추신경계 (CNS)의 병증을 완화할 수 있다는 ‘대사 싱크 (metabolic sink)’ 가설에 기초하였다. SSADH 단백질이 선천적으로 부족한 질환 마우스 모델에 SSADH mRNA-CA09 LNP를 정맥 투여하였을 때, 단회 투여로도 최소 14일 동안 전신 대사사물 수치가 정상화되는 효력을 확인하였다. 이러한 지속성은 만성 효소 결핍증 관리에서 투여 빈도를 획기적으로 낮출 수 잇는 가능성을 제시하였다.
마지막으로 제형의 안정성과 상업적 활용 가능성을 높이기 위해 최적화된 LNP 조성 (C11, 몰비 25:10:50:2)을 도출하였다. 해당 제형은 반복적인 냉해동 뿐만 아니라 영하 80도 보관 시에도 최소 6개월이상 mRNA-LNP의 물리적 성질과 기능적 활성을 유지하며 뛰어난 안정성을 보였다.
결론적으로, 본 연구는 구조-활성 관계에 기반한 이온화 지질 설계와 LNP 조성비 최적화 연구에 집중하였으며, 이를 통해 도출된 GCP LNP 플랫폼은 mRNA 기반 치료제의 전달 효율과 안전성을 동시에 확보함으로써 백신을 넘어 만성 질환을 위한 맞춤형 전달체의 가능성을 제시하였다. 이를 토대로 향후 비인간 영장류 (NHP) 모델에서의 검증과 장기 투여에 따른 안전성 평가를 거쳐, 치료 대안이 제한적인 희귀 유전질환 환자들에게 실질적인 치료 기회를 제공하는 밑거름이 될 것으로 기대된다.

목차 (Table of Contents)

• Abstract ii
• Table of Contents vi
• List of Figures x
• List of Abbreviations xiii
• Chapter 1. Introduction １
• 1.1. Study Background １
• 1.1.1. mRNA based therapeutics １
• 1.1.2. Ionizable lipid and LNP ９
• 1.1.3. mRNA １８
• 1.1.4. Focused applications: PKU and SSADH deficiency disease ２４
• 1.2. Dissertation Outline ２５
• Chapter 2. Ionizable lipid and LNP ２９
• 2.1. Introduction ２９
• 2.2. Materials and Methods ３０
• 2.2.1. mRNA synthesis ３０
• 2.2.2. Ionizable lipid synthesis ３１
• 2.2.3. mRNA-LNP formulation and characterization ３３
• 2.2.4. Cell culture and animal studies ３４
• 2.2.5. In vitro and in vivo screening of LNPs ３５
• 2.2.6. Pharmacodynamics of lead mRNA/LNPs ３５
• 2.2.7. AST/ALT measurements ３６
• 2.3. Results ３６
• 2.3.1. Design, synthesis, and characterization of a piperazine-based ionizable lipid library ３６
• 2.3.2. Formulation and physicochemical characterization of a piperazine lipid-based LNPs ４４
• 2.3.3. Structure-Activity Relationship elucidation and lead identification of a piperazine lipid-based LNPs ４８
• 2.3.4. Design, synthesis, and characterization of a cyclohexane-based ionizable lipid library ５２
• 2.3.5. Structure-Activity Relationship elucidation and lead identification of a cyclohexane lipid-based LNPs ５８
• 2.3.6. Optimization of lipid composition and identification of lead cyclohexane lipid-based LNPs ６１
• 2.3.7. Optimized GCP LNP Composition Enhances Cryostability. ６４
• 2.3.8. Pharmacokinetic and Acute Toxicity Profiling of GCP LNPs ６９
• 2.4. Discussion ７９
• Chapter 3. GCP LNPs for PKU treatment ８１
• 3.1. Introduction ８１
• 3.2. Materials and Methods ８２
• 3.2.1. mRNA synthesis ８２
• 3.2.2. mRNA-LNP formulation & characterization ８２
• 3.2.3. in vitro transfection ８２
• 3.2.4. in vivo metabolite concentration study ８４
• 3.2.5. Statistical analysis ８４
• 3.3. Results ８４
• 3.3.1. PAH protein expression and activity types based on mRNA modifications. ８４
• 3.3.2. Evaluation of PAH recovery efficiency of GCP LNPs formulated with hPAH mRNA in PKU model system ８８
• 3.4. Discussion ９１
• Chapter 4. SSADH deficiency ９３
• 4.1. Introduction ９３
• 4.2. Materials and Methods ９６
• 4.2.1. Mouse model ９６
• 4.2.2. Cell culture and fluorescent microscopic image ９６
• 4.2.3. In vitro screening of LNPs for mRNA delivery ９７
• 4.2.4. Mouse housing and experiment (SSADH) ９７
• 4.2.5. Liver sample preparation for western blot, Enzyme linked immunosorbent assay (ELISA) and activity assay ９７
• 4.2.6. Western blot ９８
• 4.2.7. ELISA ９８
• 4.2.8. Activity assay ９９
• 4.2.9. LC-MS/MS for GHB and GABA quantification １００
• 4.3. Results １０１
• 4.3.1. mRNA Selection and Comparison １０１
• 4.2.2. Evaluation of mRNA/LNP-Mediated SSADH Enzyme Expression and Functional Impact in SSADH Knock-Out Mice １０４
• 4.2.3. Comparative Evaluation of Optimized LNP Formulations and Nucleoside-Modified mRNAs in SSADH Knock-Out Mice １０７
• 4.4. Discussion １１０
• Chapter 5. Conclusion and Future Work １１２
• Bibliography １１４
• Abstract in Korean １３４