In vivo and in situ delivery of biotherapeutics for improved extrahepatic bioavailability

<jats:p>&lt;p dir="ltr"&gt;Biotherapeutics have transformed the treatment of several diseases, yet their clinical application remains largely confined to hepatic indications. The constraint is anatomical rather than chemical: the liver receives approximately a quarter of cardiac output through a fenestrated, discontinuous sinusoidal endothelium that grants systemically administered macromolecules and nanoparticles direct access to a receptor-dense parenchyma. Achieving comparable delivery efficiency in extrahepatic tissues therefore remains a central challange. This thesis investigates two strategies to improve the bioavailability of biotherapeutics beyond the liver: covalent lipid conjugation of antisense oligonucleotides (ASOs), and endogenous engineering of extracellular vesicles (EVs) for intracellular cargo delivery, including their production within the body itself.&lt;/p&gt;&lt;p dir="ltr"&gt;ASOs are short synthetic nucleic acid therapeutics that bind a target RNA by Watson-Crick base pairing to silence its expression, offering sequence- programmable specificity but poor delivery beyond the liver. Paper I examined how lipid conjugation influences the biodistribution and activity of phosphorothioate gapmer ASOs. After subcutaneous administration, fatty acid conjugation improved target knockdown in heart and kidney but also enhanced hepatic silencing, indicating that lipid conjugation acts as a pharmacokinetic potentiator rather than a targeting mechanism; cholesterol conjugation restricted activity largely to the liver and reduced potency. Single-cell RNA sequencing resolved uptake to liver sinusoidal endothelial cells and hepatocytes, and revealed transcriptional responses attributable to the lipid moiety itself. Following intracerebroventricular administration, a route not previously examined for these chemistries, fatty acid conjugates improved silencing in deep brain regions whereas cholesterol conjugation reduced delivery throughout the CNS, remaining confined to the ventricular system as evidenced by imaging of cleared tissue. We propose that the approximately 150-fold lower protein concentration of cerebrospinal fluid relative to plasma saturates available lipid-binding carriers, leaving unbound conjugate to associate with lipid-rich ependymal surfaces. Lipophilicity that confers an advantage in the circulation thus becomes a liability in the CSF, establishing route of injection as a critical determinant of conjugate performance.&lt;/p&gt;&lt;p dir="ltr"&gt;EVs are cell-derived membrane nanoparticles that naturally transfer molecular cargo between cells, making them attractive vehicles for intracellular delivery of therapeutics that cannot otherwise cross the cell membrane. Realising this potential requires loading a chosen cargo into the vesicle and releasing it into the target-cell cytosol, the two barriers Paper II addressed. A self-cleaving mini-intein inserted between an EV-sorting domain and the cargo released soluble cargo into the vesicle lumen during biogenesis, overcoming the constraint that membrane- tethered cargo cannot reach its site of action. A screen of 40 human-derived and two viral fusogenic proteins identified VSV-G as the only candidate conferring efficient endosomal escape. Combining active loading, intein-mediated release, and VSV-G-driven escape enabled functional intracellular delivery of Cre recombinase, Cas9/sgRNA RNP complexes, and a meganuclease. In vivo, intracerebroventricular infusion of Cre-loaded EVs produced recombination in over 40% and 30% of hippocampal and cortical cells respectively, and functional delivery was further demonstrated following intratumoral injection and systemic administration. Systemic delivery of engineered EVs carrying an NF-KB super- repressor significantly improved survival in a model of LPS-induced systemic inflammation.&lt;/p&gt;&lt;p dir="ltr"&gt;Paper III evaluated the contribution of the EV-sorting scaffold to the cargo loading efficiency. Three tetraspanins outperformed the canonical CD63 scaffold, with TSPAN2 the most effective; single-vesicle analysis revealed substantially higher co-display of the sorting domain and VSV-G for TSPAN2, suggesting that scaffold- dependent differences in biogenesis govern fusogen co-incorporation and thus delivery per vesicle. TSPAN2-engineered EVs achieved recombination in melanoma xenografts and in primary cells from mice, and retained therapeutic efficacy in the LPS- induced systemic inflammation mouse model.&lt;/p&gt;&lt;p dir="ltr"&gt;Paper IV applied this endogenous loading principle to EVs produced within the body, circumventing ex vivo manufacture and repeated bolus dosing. Across clinically relevant gene delivery modalities, surface display of an albumin-binding domain extended circulatory residence of EVs in blood and tethering cargo to the EV scaffold protein CD63 packaged it into secreted vesicles that carried it out of the liver, where it is produced, to extrahepatic tissues, whereas untethered cargo remained largely hepatic. Functional cargo transfer by in situ-produced EVs was demonstrated in reporter mice. Coupling production of engineered EVs to an NF- KB-responsive promoter established a self-regulating biofactory governed by the inflammatory microenvironment; therapeutically, however, constitutively produced EVs outperformed the inducible system, consistent with the kinetic advantage of a pre-existing vesicle pool at the onset of inflammation. Across modalities gene delivery modalities, EV-packaged cargo showed greater relative extrahepatic distribution than soluble controls.&lt;/p&gt;&lt;p dir="ltr"&gt;Together, these studies indicate that extrahepatic delivery is achieved not by any single modification but by layering design choices whose effects depend strongly on administration route and biological compartment. The studies further suggest that hepatic tropism doesn't need to be circumvented but rather turned into advantage to produce a therapeutic in liver that acts elsewhere in the body. This principle reframes field's major obstacle as a delivery mechanism.&lt;/p&gt;&lt;h3 dir="ltr"&gt;List of scientific papers&lt;/h3&gt;&lt;p dir="ltr"&gt;I. &lt;b&gt;Roudi S*&lt;/b&gt;, Estupiñán* HY, Saher O, Barradas C, Inganäs E, Frengen N, Grochowski R, Pavlova S, Nordin JZ, Biscans A, Matsson P, Zain R, Sandberg R, Hagemann-Jensen M, El Andaloussi S. Lipid-ASO therapeutics exhibit differential tissue targeted delivery upon systemic or local CNS administration. [Manuscript]&lt;/p&gt;&lt;p dir="ltr"&gt;II. Liang X, Gupta D, Xie J, Van Wonterghem E, Van Hoecke L, Hean J, Niu Z, Ghaeidamini M, Wiklander OPB, Zheng W, Wiklander RJ, He R, Mamand DR, Bost J, Zhou G, Zhou H, &lt;b&gt;Roudi S,&lt;/b&gt; Estupiñán HY, Radler J, Zickler AM, Görgens A, Hou VWQ, Slovak R, Hagey DW, de Jong OG, Uy AG, Zong Y, Mager I, Perez CM, Roberts TC, Carter D, Vader P, Esbjörner EK, de Fougerolles A, Wood MJA, Vandenbroucke RE, Nordin JZ, El Andaloussi S (2025). Engineering of extracellular vesicles for efficient intracellular delivery of multimodal therapeutics including genome editors. Nat Commun 16(1), 4028. &lt;a href="https://doi.org/10.1038/s41467-025-59377-y" target="_blank" rel="noreferrer"&gt;https://doi.org/10.1038/s41467-025-59377-y&lt;/a&gt;&lt;/p&gt;&lt;p dir="ltr"&gt;III. Niu Z, Zhou H, Zheng W, Hayes OG, Hou VWQ, Görgens A, &lt;b&gt;Roudi S,&lt;/b&gt; Zhou G, Wiklander RJ, Sych T, Sezgin E, Nordin JZ, Zhao Y, Liang X, Andaloussi SEL (2025). Screening scaffold proteins for improved functional delivery of luminal proteins using engineered extracellular vesicles. J Control Release, 10(384):113882. &lt;a href="https://doi.org/10.1016/j.jconrel.2025.113882" target="_blank" rel="noreferrer"&gt;https://doi.org/10.1016/j.jconrel.2025.113882&lt;/a&gt;&lt;/p&gt;&lt;p dir="ltr"&gt;IV. &lt;b&gt;Roudi S,&lt;/b&gt; Görgens A, Estupiñán HY, Barradas C, Liang X, Rädler J, Zhou H, Mowoe M, Zickler AM, Makkar R, Karatheodorou E, van Den Heuvel J, Huang Y, Hou W, Nordin JZ, Zheng W, Gupta D*, EL Andaloussi S *. In situ production of engineered extracellular vesicles for efficient delivery of protein biotherapeutics. [Manuscript]&lt;/p&gt;&lt;p dir="ltr"&gt;*These authors contributed equally.&lt;/p&gt;</jats:p>