New mRNA Nanoparticle Delivers Dual Fat-Burning Hormones Directly to Adipose Tissue

Obesity drives type 2 diabetes, cardiovascular disease, and metabolic-dysfunction-associated steatotic liver disease on a global scale. While protein-based drugs like semaglutide (GLP-1 receptor agonist) and FGF21 analogs show meaningful efficacy, they require frequent injections, suffer from proteolytic instability, and lack tissue specificity. mRNA therapeutics offer an alternative by enabling transient, in situ protein production — but delivering mRNA effectively to fat tissue rather than the liver has remained a major unsolved challenge. This study from Peking University's School of Pharmaceutical Sciences tackles that problem with a rationally designed lipid chemistry platform and a potent dual-hormone payload.

The team synthesized a 24-member library of mono-amine phosphoramidate lipids using a two-step nucleophilic substitution of phosphorus oxychloride. Two structural classes emerged: two-tailed phosphoramidate lipids (PL) and three-tailed N-alkyl phosphoramidate lipids (NPL), where an additional hydrophobic alkyl chain was conjugated to the amine nitrogen. After formulating all candidates with standard helper lipids (DOPE, cholesterol, DMG-PEG2000) and firefly luciferase mRNA, each was injected subcutaneously into the inguinal fat region of C57BL/6J mice and evaluated via IVIS bioluminescence imaging at 6 hours. Across the entire library, N-alkyl modifications consistently outperformed their two-tailed counterparts, and NPL20 — featuring a 12-carbon N-alkyl chain and branched O-alkyl tails — emerged as the lead candidate.

NPL20 LNPs accumulated approximately 3.4-fold more in adipose tissue than PL16 LNPs and about 1.4-fold more than NPL19 and PL17. More strikingly, mRNA expression was nearly 12-fold greater than the average of the comparison lipids, indicating that both enhanced cellular uptake and superior endosomal escape contribute to NPL20's efficiency. The lipid's pKa of 7.5–8.0 was identified as critical: it facilitated electrostatic interaction with glycosaminoglycans overexpressed on adipocyte surfaces, enabling local retention without the immunogenicity of highly cationic lipids (pKa >8) or the lymph-node and liver drainage typical of neutral LNPs (pKa 6.3–6.5). Membrane fluidity studies and galactose competition assays confirmed that NPL20's performance relied on both pKa-mediated surface binding and enhanced membrane fusion capacity.

A key formulation discovery was that removing DOPE entirely — while retaining cholesterol — boosted delivery efficiency approximately 5-fold compared to the conventional four-component LNP, creating a simpler three-component system. This simplified NPL20 formulation avoided liver stress markers (ALT/AST) and inflammatory cytokine elevation (TNF-α, IL-6) after repeated dosing, an important safety advantage. The therapeutic mRNA payload encoded a GLP-1/FGF21 fusion protein stabilized by an IgG4 Fc domain (mGLP-1/FGF21-Fc), which extended half-life via FcRn-mediated recycling. An alternative VLK-tagged construct for albumin binding was also tested. In diet-induced obese C57BL/6J mice, treatment with mGLP-1/FGF21-Fc LNPs produced significant reductions in body weight and total fat mass, preserved lean mass, lowered fasting glucose and insulin levels, improved glucose tolerance, and substantially reduced hepatic lipid accumulation — demonstrating synergistic action through GLP-1's appetite suppression and FGF21's thermogenic and lipolytic effects.

This platform represents a meaningful advance in the field of mRNA-based metabolic therapy. By engineering adipose-tropic nanoparticles rather than relying on systemic or hepatic delivery, the approach localizes therapeutic protein production to the tissue most relevant to fat metabolism. The dual-agonist design addresses a limitation of single-hormone therapies, and the IgG4 Fc stabilization strategy reduces dosing frequency. Caveats include the preclinical nature of the work (mice only), uncertainty about translation of subcutaneous adipose targeting to human anatomy, and the absence of long-term toxicology or primate data.