(W1230-07-45) Comprehensive Design and Characterization of pH-Gradient-Loaded Donkey-Milk Exosomes for Oral Octreotide Delivery: A Bench-to-In Silico Roadmap

Purpose: Octreotide (OCT) is confined to intramuscular injection because it is rapidly degraded and poorly absorbed in the gastrointestinal tract. We aimed to create the first orally deliverable OCT formulation by harnessing donkey-milk exosomes naturally enriched in membrane-fusion and transcytosis proteins to shield the peptide from gastric degradation and transport it across intestinal epithelium. Our strategy combined (i) a mild pH-gradient loading protocol to entrap OCT within the vesicle lumen, (ii) systematic in-vitro and MDCK permeability testing, (iii) forward-predicted in-silico PBPK simulations with GastroPlus™, and (iv) in-vivo pharmacokinetic and tissue-distribution studies in mice, establishing an integrated pipeline for converting injectable peptides into effective oral therapeutics.

Methods: Exosomes were isolated from donkey milk powder (Dulcededonke, OK, USA) using sodium citrate precipitation followed by ultra-centrifugation method and thoroughly characterized by (a) nanoparticle tracking analysis (NTA and fluorescence-NTA), (b) total protein quantification, (c) western blotting for exosomal markers (Alix, CD61, CD83, HSP70), and (d) proteomic profiling using LC-MS/MS Analysis (Brukers, timsTOFHT, Metware Biotechnology Inc.). To overcome OCT’s poor (~1%) oral bioavailability, the peptide was entrapped in these exosomes via a pH-gradient loading technique using acidic pH 4.6 and neutral pH 7.4 conditions. The acidic pH exosomes were incubated for 2 hrs with OCT dissolved in neutral pH conditions (pH 7.4) and characterized with various analytical tools. The OCT-loaded exosomes then underwent in-vitro dissolution testing under simulated gastric (pH 1.2) and intestinal (pH 6.8) conditions at 37°C temperature, followed by an absorptive permeability assay across MDCK cell monolayers (TEER >500 Ω·cm²), with permeability quantified by a validated analytical method. The in-silico GastroPlus™ simulations tools predict the fraction absorbed (Fa), Tmax, and AUC for both pure OCT and OCT loaded exosomes. Finally, pharmacokinetic studies in male BALB/c mice (2 mg/kg oral dose) conducted to compare the plasma exposure (AUC₀–₂₄) and tissue distribution of free versus exosome-encapsulated OCT.

Results: Donkey milk derived exosomes exhibited a mean particle size of 136.2 ± 0.65 nm, a zeta potential of -33.3 ± 0.23 mV, and a protein concentration of 2.5 mg/mL (Figure 1a, 1b). Western blotting confirmed the presence of exosomal markers Alix, CD61, CD83, and HSP70, while fluorescent NTA demonstrated that 92 % of particles incorporated the fluorescent probe (Figure 1c, 1d). Proteomic analysis revealed abundant transport-facilitating proteins; polymeric immunoglobulin receptor (PIGR; A0A8C4L7D7), milk fat globule–EGF factor 8 (MFGE8; A0A8C4MDY3, A0A8C4MFR8), annexin A2 (ANXA2; A0A8C4LKS4) along with tetraspanins CD9 (A0A8C4MJ30), CD63 (A0A8C4KX98), and CD81 (A0A8C4KW70), all of which are implicated in enhanced oral uptake (Figure 1e). OCT was loaded into the exosomes via a pH-gradient technique where the acidified extracellular vesicles were incubated with the peptide at pH 7.4, yielding an entrapment efficiency of 7.25 ± 0.23 % (LC-MS). In-vitro dissolution in phosphate-buffered saline (pH 7.4) and simulated gastric fluid (pH 1.2) at 37°C showed that, after 24 h at pH 6.8, 55±2.27 % of free OCT was released versus 40±3.15 % from the exosome-encapsulated formulation (Figure 2a, 2b). MDCK monolayer permeability assays (TEER > 500 Ω·cm²) measured an apparent permeability coefficient (Papp) of 17.26 × 10⁻⁶ ± 1.42 cm/s for OCT-loaded exosomes versus 0.46 × 10⁻⁶ ± 0.09 cm/s for free peptide (Figure 1e). It was observed that the apparent permeability of OCT from OCT exosomes was improved by 37.52-folds (P < 0.0001) in comparison to OCT control. GastroPlus™ simulations predicted a rise in fraction absorbed (Fa) from 8.5 % to 88.2 %, a reduction in Tmax from 5.5 h to 2.2 h, and a ~4-fold increase in AUC₀–t (0.0025 → 0.0098 µg·h/mL) (Figure 2c, 2d). In-vivo PK studies showed enhanced bioavailability of OCT from OCT-exosomes with AUC(0-24h) of 435200 ± 17700 pg·h/mL as compared to OCT control with AUC(0-24h) of 27153.76 ± 4788.40 pg·h/mL with a single oral dose of 2 mg/kg (Figure 3a). The data suggests that Donkey milk exosomes significantly improved the oral bioavailability of OCT. The biodistribution of OCT in various tissues results in confirmation of OCT in liver, spleen, kidney, and intestine tissues (Figure 3b, 3c).

Conclusion: This first‐in‐field demonstration, we show that donkey-milk exosomes naturally enriched with mucosal-transcytosis proteins can transport octreotide across the gut, converting a strictly injectable peptide into an effective oral therapy. A single-step pH-gradient method secured stable cargo loading (~7.25%), boosted MDCK permeability 37-fold, and, as predicted by GastroPlus™, translated to a 16-fold rise in mouse AUC with targeted distribution to liver, spleen, kidney, and intestine. This platform is immediately scalable, offering a practical, broadly applicable route for oral delivery of other fragile biologics.


Figure 1. Nanoparticle tracking analysis and Western blotting of exosomes. a) Particle size distribution of Donkey milk exosomes; b) Zeta potential of Donkey milk exosomes; c) Size of Donkey milk exosomes in scatter and fluorescence mode; d) Western blots analysis of CMDE shows expression of proteins Alix, CD63, CD81 and HSP 70; e) Apparent permeability coefficient (Papp) of OCT and OCT-Exosomes in MDCK monolayers. Results were expressed as Mean ± SD (n = 3). ****P < 0.0001; e) List of significant proteins in the proteomics analysis


Figure 2 In-vitro dissolution study at a) pH 1.2 buffer showing cumulative drug release from OCT and OCT loaded exosomes; b) pH 6.8 buffer showing significantly enhanced drug release from OCT loaded exosomes as compared to OCT; c) In-silico pharmacokinetic analysis of free OCT vs OCT loaded exosomes. OCT absorption in each segment of the GIT; d) Predicted mean plasma concentration of OCT absorbed from free OCT and OCT loaded exosomes.


Figure 3 a) Pharmacokinetic study in BALB/c mice showing AUC (0–24) and Cmax in OCT loaded exosomes and OCT control; b) Bio-distribution studies of OCT from OCT-exosomes in various tissues; c) Bio-distribution studies of OCT from control. Results were expressed as Mean ± SD (n = 4).