Immunoregulatory protein-hybrid extracellular vesicles via self-loadable backbone cyclization for oral inflammatory bowel disease therapy

Authors:

Abstract
Despite the potential of extracellular vesicles (EVs) and therapeutic proteins as carriers and cargos for oral delivery, their precise functional integration remains a persistent challenge, thereby limiting synergistic therapeutic outcomes. Here, we present a protein-hybrid, orally delivered EV strategy that integrates naturally bioactive EVs with self-loadable, backbone-cyclized immunoregulatory proteins for inflammatory bowel disease (IBD) therapy. Through computationally guided design and split intein-mediated backbone cyclization, we generated cyclized variants of key immunoregulatory proteins with improved functionality. Among these, C-R4-tagged cyclization of phosphatase domain of T-cell protein tyrosine phosphatase (ppTCPTP) improved membrane permeability, thermal stability, and anti-inflammatory activity. This backbone cyclization enabled efficient and high-capacity loading of ppTCPTP into native EVs that are not amenable to genetic engineering. Notably, these protein-hybrid EVs exhibited acidic resistance for oral delivery and synergistically enhanced antioxidant and anti-inflammatory effects in murine IBD organoids and in vivo colitis models, markedly reducing intestinal inflammation and restoring epithelial barrier integrity. Our findings highlight the translational potential of this self-loadable protein-EV platform as a safe and potent oral biologic for IBD therapy.
Funding Information
  • Ministry of Science and ICT, South Korea (RS-2025-00553158)
  • Ministry of Science and ICT, South Korea (RS-2025-02232977)
  • Ministry of Science and ICT, South Korea (KFRM24A0105L1)
  • Ministry of Health and Welfare (KFRM25A0107L1)
  • Ministry of SMEs and Startups (RS-2023-00282845)
  • Korea Basic Science Institute (2023R1A6C101A009)
  • National Research Foundation of Korea (2020R1A6A1A06046728)
  • Ministry of Education

This publication has 64 references indexed in Scilit: