Dynamic Covalent Disulfide Exchange Mediates Oral Delivery of Biomacromolecules

Dynamic Covalent Disulfide Exchange Mediates Oral Delivery of Biomacromolecules | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Biological Sciences - Article Dynamic Covalent Disulfide Exchange Mediates Oral Delivery of Biomacromolecules Huanghao Yang, Chen Chen, Tingjing Huang, Zheng Liu, Hongjun Li, and 12 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3616020/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The biological barriers present in the intestine thwart the absorption of orally delivered biologics, which, if overcome, would reduce the injection burdens for millions of patients. Here, we present a straightforward yet effective oral biologic formulation, which utilizes in-situ growing poly(disulfide)s as the sole excipient to circumvent all intestinal barriers in a noninvasive way. We find that, through dynamic covalent disulfide exchange initiated by the thiols in mucins, epithelial membranes, and hepatic sinusoids, digestion-resistant complex coacervates formed from insulin (as a model drug) and guanidinium-containing poly(disulfide)s readily traverse the mucus and epithelial layers without remodeling the barriers’ integrity, and then undergo dissociation to release insulin in the liver. Oral gavage of the complex coacervates with enteric capsules into diabetic mice and swine models elicited a-few-hour longer hypoglycemic effect than subcutaneously injected insulin, attaining relative bioavailabilities over 20%. 14-day dosing experiments demonstrated the postprandial and daily glycemic control capacity and the biosafety of this oral insulin. The generality of this formulation scheme was further validated with human serum albumin and salmon calcitonin. Biological sciences/Drug discovery/Drug delivery Health sciences/Medical research/Drug development drug delivery oral biologics dynamic covalent chemistry poly(disulfide)s complex coacervates diabetes treatment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Oral biologics have been pursued since the introduction of insulin injection for diabetes mellitus management 100 years ago. 1-3 Compared to injection, oral administration is often preferred due to its convenience, painlessness, and superior adherence. 4-6 For insulin in particular, oral delivery could closely recapitulate the physiological route of endogenous insulin delivery from the pancreas to the liver via the hepatic portal circulation, obviate peripheral hyperinsulinemia, and allow for earlier initiation of insulin therapy. 6,7 However, the implementation of oral biologics has been constrained by the inherent physiological function and anatomical structure of the gastrointestinal tract. While the acidic milieu in the stomach has been circumvented by technologies like enteric coatings, the formidable barriers in the intestine, including digestive enzymes, the mucus lining, and the epithelial layer, fundamentally impede the absorption of undegraded biomacromolecules. 8-10 Substantial efforts have been attempted to navigate these hurdles via co-formulating biologic drugs with multiple ingredients (including vehicles, enzyme inhibitors, mucus-penetrating agents, and permeation enhancers) or loading biomacromolecules into swallowable miniature devices. 11-19 Nonetheless, these innovations inevitably increase formulation complexity and raise biological barrier manipulation associated biosafety concerns such as bowel infections, perforation, and obstruction. 6,8 We here describe a facile oral biologic formulation which utilizes in-situ growing poly(disulfide)s as the sole excipient but can surmount all intestinal barriers without compromising their integrities. Poly(disulfide)s, typically grown from dithiolanes such as lipoic acid derivatives, have shown to mediate the translocation of cargoes across the cell membrane into the cytosol. 20-23 The crux of this cellular entry paradigm is the dynamic covalent disulfide exchange (DCDE) chemistry which begins with exofacial thiols and continues along sulfur tracks in transmembrane proteins and can be inhibited by thiol-blocking reagents. 24-26 DCDE is also essential in biology for the maintenance of cellular redox homeostasis, protein folding, and virus entry. 24-26 The magic of this natural dynamic covalent chemistry, especially for mediating systemic delivery of cargoes, is far from the limit. In this study, focusing on insulin coacervated with poly(disulfide)s grown in situ from lipoic acid-guanidinium derivatives (LGs; Fig. 1 ), we set out to establish the breadth and mechanism of DCDE in facilitating oral delivery of biomacromolecules, and to evaluate the pharmacodynamics, pharmacokinetics, glycemic controllability, and biosafety of this oral insulin in diabetic mice and swine models. Additionally, to showcase the modularity of this formulation strategy, another two oral biologics, human serum albumin (HSA) and salmon calcitonin (sCT), were confected, and tested in murine models. Results Complex coacervation of insulin with in-situ growing poly(disulfide)s Given that dithiolanes tend to undergo concentration-dependent, strain-release-promoted ring-opening polymerization 20,21 , we surmise that, when the negatively charged patches on insulin surface bind the positively charged heads of amphiphilic LGs via salt bridges, the resulting decrease in net surface charge together with an increase in hydrophobicity will induce preliminary phase separation 27 , whereby the locally concentrated dithiolanes spontaneously grow into poly(disulfide)s. The latter concomitantly form multivalent interaction-stabilized complex coacervates with insulin ( Fig. 1a, b, Supplementary Fig. 1 ). To test this hypothesis, we mixed LGs ( Supplementary Figs. 2-8 ) and insulin at different molar ratios (LG/insulin = x ) and observed that all solutions instantly turned turbid ( Supplementary Fig. 9 ). Meanwhile, the characteristic absorbance of dithiolanes at λ = 330 nm gradually declined and reached a plateau within 15 min ( Supplementary Fig. 10 ), while peaks of oligomeric poly(disulfide)s corresponding to n = 2–9 were revealed by MALDI-TOF analysis ( Supplementary Fig. 11 ), validating the occurrence of ring-opening polymerization after preliminary condensation. Cryogenic electron microscopy and dynamic light scattering showed that nanosized complex coacervates were formed ( Fig. 1c, Supplementary Fig. 12 ). In contrast, in control experiments performed at pH 4.0 (below the isoelectric point of insulin) or when using insulin plus guanidinium chloride, insulin plus lipoic acids, or the individual ingredients, neither ring opening nor complex coacervation occurred ( Supplementary Figs. 9, 10, 12 ), further validating our formulation scheme. As x increased from 10 to 80, the average size, surface charge, and encapsulation efficiency of complex coacervates (termed LG/insulin- x ) initially increased and then plateaued at x above 40, while the loading capacities showed the opposite trend ( Fig. 1d, Supplementary Figs. 12, 13 ). Moreover, the secondary structure of insulin was preserved ( Supplementary Fig. 14 ), indicating that the conserved disulfide bridges of insulin were not disturbed by the ring-opening polymerization. DCDE-mediated transepithelial transport As the epithelial layer is regarded as the rate-limiting barrier for drug absorption, we next assessed the capacity of LG/insulin- x to traverse Caco-2 monolayers ( Supplementary Fig. 15 ), a common in vitro model for the intestinal epithelium. 28 In this section, studies were continued with LG/insulin-10, -20, and -40. LG/insulin-40 achieved the fastest transepithelial delivery, reaching 48.4% within 6 h, and yieldedthe highest apparent permeability coefficient ( P app ) of 17.8×10 -6 cm s -1 , which is approximately 50.6, 4.6 and 1.9 times greater than with free insulin, LG/insulin-10, and LG/insulin-20, respectively ( Fig. 2a, b ). Although sodium caprate (an intestinal permeation enhancer used in clinical trials 29 ) produced a similar insulin transport profile to LG/insulin-40, the integrity of the tight junctions of Caco-2 monolayers was severely compromised, which was not the case with LG/insulin- x ( Supplementary Figs. 16, 17 ). Meanwhile, MTT and lactate dehydrogenase release assays indicated that LG/insulin- x exerted trifling influence on the cell viability or membrane permeability ( Supplementary Fig . 18 ). Together, these results suggest that the transepithelial transport of LG/insulin-x occurs through transcellular pathways. To gain more in-depth insights into the mechanism, we first studied the effect of classical inhibitors of endocytosis and exocytosis. These inhibitors slightly attenuated the permeation of LG/insulin- x ( Fig. 2c, Supplementary Fig . 19 ), suggesting that regular transcytosis pathways marginally mediated this transcellular progression. However, extracellularly added or fusogenic liposome-delivered thiol-blocking agent (5,5ʹ-dithio-bis-(2-nitrobenzoic acid), DTNB; Supplementary Fig . 20 ) dramatically slowed the passage of LG/insulin- x , implying that the DCDEC between the exofacial/intracellular thiols and the disulfides in LG/insulin- x was decisive for mediating this underexplored transcellular transport. We then tracked the cellular uptake and subcellar localization of LG/insulin- x over time. Following incubation for 10 min, we saw the insertion of LG/insulin- x in the plasma membrane and this binding was curtailed by DTNB but not endocytosis inhibitors ( Fig. 2d, Supplementary Figs. 21-24 ). Moreover, a local increase in lipid order ( Fig. 2d ) and reversible flip-flop of lipids ( Supplementary Fig. 25 ) were observed in the binding region. Meanwhile, the non-influx of the cell impermeant propidium iodide suggested that the membrane remained intact throughout. As such adaptive membrane structural changes are reminiscent of those observed in the formation of transient micellar pores by cell-penetrating peptides when they mediate the uptake of cargoes, the translocation of LG/insulin- x might follow a similar mechanism except that here the DCDE between the disulfides of LG/insulin- x and transmembrane sulfur networks facilitated this process ( Fig. 2e ). 24,30,31 Over time, LG/insulin- x was primarily delivered to the cytosol and its internalization was dramatically reduced by extracellularly added DTNB, but was insensitive to endocytosis inhibitors ( Supplementary Figs. 26-30 ). Analysis of the correlations between LG/insulin- x and organelles, including endo-lysosomes, Golgi apparatus, endoplasmic reticulum, mitochondria, and microtubules, revealed no specific colocalizations. The cytosolic delivery paradigm of LG/insulin- x is typical for DCDE-mediated cellular uptake 21-23 , which would help to avoid degradation associated with the endo-lysosomal endocytosis pathway. As the subcellular distribution pattern was not affected by endo-/exocytosis inhibitors, the movement of LG/insulin- x across the cytosol to the basal membrane was less likely mediated by transcytosis machineries, but perhaps driven by thermal motions or aggregate cytoplasmic forces which were suggested to exert on cytosolic delivered cargoes 32,33 . Further, confocal x-z sections revealed that the basal membranes was stained by the intracellularly delivered LG/insulin- x , as observed during early uptake, and this binding was affected by intracellularly delivered DTNB but not exocytosis inhibitors ( Fig. 2f, Supplementary Fig. 31 ). These findings suggest that the egression of LG/insulin- x across the basal membrane was a similar but inverse process of thiol-mediated uptake, relying on the DCDE initiated by endofacial thiols. Additionally, we noted that most LG/insulin-40 remained as discrete puncta in the cytosol, while diffuse fluorescence spreading throughout the cytoplasm was observed for LG/insulin-10 and -20 over time ( Supplementary Figs. 26-30 ). Another finding was that intracellularly delivered DTNB reduced this diffuse fluorescence to a certain extent. These phenomena hinted at the sensitivity of LG/insulin- x to the cytosolic glutathione, which could trigger reductive depolymerization of poly(disulfide)s, leading to the dissociation of complex coacervates and insulin leakage. To further understand the mechanism, we studied the dissociation kinetics of LG/insulin- x in solutions containing different concentrations of glutathione. Indeed, LG/insulin-40 disassembled much slower than the other two formulations at the cytosolic glutathione level of 0.3–1.6 mM in Caco-2 cells 34 ( Supplementary Fig. 32 ). The higher stability of LG/insulin-40 stemmed from the higher content and longer chain length of poly(disulfide)s ( Fig. 1d, Supplementary Fig. 11 ), which provided greater cross-linkages with insulin to conceal themselves better against glutathione cleavage 35,36 , thereby imparting superior transcellular delivery efficacy. Based on these preliminary experiments, LG/insulin-40 was leveraged for further studies, presented below. Stability against digestion and transmucosal delivery We next examined the potency of LG/insulin-40 (20 IU kg -1 ) for transmucosal delivery in vivo by intrajejunal injection to diabetic mice, which was readily assessed by monitoring changes in blood glucose levels (BGLs). After intrajejunal injection, BGLs steadily decreased and reached a maximum reduction of approximately 70% within 2 h ( Fig. 3a, Supplementary Fig. 33 ). BGLs remained below 200 mg dl -1 for nearly 7 hours, which is about 2.5-hour longer than with subcutaneous (s.c.) insulin (5 IU kg -1 ). Integration of the areas above the BGL curves (AACs) yielded a pharmacological availability of 27.5% ( Fig. 3b ). However, intrajejunal injection of free insulin did not affect BGLs. Although intrajejunal injection of sodium caprate plus insulin produced certain BGL-lowering response, the relative bioactivity was about 2.3-times lower than that of LG/insulin-40. Importantly, unlike sodium caprate which caused severe intestinal tight junction opening, LG/insulin-40 had no influence on the intestinal tissue morphology which remained intact like in the untreated control ( Fig. 3c ). Such a high bioactivity produced by this simple formulation is surprising given the presence of two other physiological barriers in the intestinal tract compared to the Caco-2 monolayer model, i.e. , digestive enzymes and mucus. The stability of LG/insulin-40 against proteolytic degradation was studied in simulated intestinal fluid. Compared to free insulin, which was fully degraded within 1.5 h, approximately 80% of insulin in LG/insulin-40 was preserved over 6 h incubation and the suspension of LG/insulin-40 remained turbid still, indicating that the complexation of insulin and poly(disulfide)s into stable coacervates protected insulin against proteolytic degradation in the intestine ( Fig. 3d, Supplementary Fig. 34 ). We then sought to answer how LG/insulin-40 migrated through the mucus by using a 3D migration model reconstituted with Type II mucin. We found that LG/insulin-40 homogeneously adhered to the mucus, travelled at least 200 μm within 15 min, and achieved 57.4% cumulative transport over 6 h with a P app value of 24.9×10 -6 cm s -1 ( Figs. 3e-g ). However, in experiments conducted with DTNB-pretreated mucus, most of LG/insulin-40 spread beyond the area where it was dropped, and the adhered LG/insulin-40 showed a shallower penetration depth. Additionally, free insulin exhibited limited binding to and penetration into the mucus. The pronounced mucus adhesion and permeation capacity of LG/insulin-40 was unlikely achieved via mucus thinning or lysis, since there was neither drop in mucus viscosity nor cleavage of mucins ( Supplementary Figs. 35, 36 ). Given the cysteine-rich nature of mucins 37,38 , it is reasonable to deduce that their sulfurs can provide dynamic covalent exchange networks to transiently capture LG/insulin-40 and promote its accumulation in the mucus, which generates concentration gradients between the upper and lower mucus layers to drive the diffusion of LG/insulin-40 towards the epithelia ( Fig. 3h ). Oral insulin delivery in diabetic mice We next assessed the capacity of LG/insulin-40 enabling oral delivery by gavage of diabetic mice using concentrate-loaded enteric capsules (size M, 20 IU kg -1 ; Supplementary Fig. 37 ). Oral ingestion of LG/insulin-40 solution (20 IU kg -1 ), free insulin-loaded capsules (20 IU kg -1 ) or blank capsules, and s.c. insulin (5 IU kg -1 ) were adopted as controls. Unlike s.c. insulin, which induced a sharp drop but a quick recovery in BGLs, oral administration of LG/insulin-40-loaded capsules provoked much longer hypoglycemic effect with a relative bioactivity of 26.7% ( Fig. 4a, b, Supplementary Fig. 38 ). Oral gavage of LG/insulin-40 solution did not elicit obvious drops in BGLs, due to its swift dissociation and degradation in harsh gastric milieu ( Supplementary Fig. 39 ). Correspondingly, for mice receiving s.c. insulin, the plasma insulin concentration rapidly increased within 1 h, but then decreased in the following 2 h ( Fig. 4c ). In contrast, the plasma exposure of insulin produced by LG/insulin-40-loaded capsules peaked at 4 h and was detectable over 8 h postdosing, yielding a relative oral bioavailability of 28.3% ( Fig. 4d ). For groups treated with LG/insulin-40 solution, or blank or free insulin-loaded capsules, BGLs and plasma insulin levels remained basal throughout. We then tracked the passage of LG/insulin-40, where insulin was pre-tagged with Cy5.5 (insulin Cy5.5 ), in the intestinal tract. After oral gavage of LG/insulin Cy5.5 -40-loaded capsules, fluorescence signals propagated from the duodenum to jejunum, ileum, and colon over time, while the intensity of each segment firstly increased and then gradually weakened ( Fig. 4e ). Cryosections of each lighted segment revealed prominent fluorescence signals inside the villi, evidencing that LG/insulin-40 crossed the mucosa irrespective of the intestinal regions. The permeation efficiency of LG/insulin-40 across different intestinal regions were evaluated ex vivo using an Ussing chamber system, which yielded P app values in the same order of magnitude (10 –4 ) without significant differences ( Supplementary Fig. 40 ). Addition of DTNB or DTNB-loaded fusogenic liposomes in the donor chamber significantly curtailed the absorptive transport of LG/insulin-40 ( Supplementary Fig. 41 ). These results demonstrated that the intestinal uptake of LG/insulin-40 hinged on the DCDE initiated by the thiols in the mucosa, while no obvious preferential absorption was biased towards one segment. We also studied the distribution of signals of insulin Cy5.5 in major organs ( Fig. 4f ), which revealed that fluorescence signals appeared in the liver 1 h after oral dosing and reached the maximum after 4 h. At each post-gavage time point, the fluorescence intensity was relatively stronger in the liver, followed by kidneys, but remained low in the heart, spleen, and lungs. This biodistribution pattern was concordant with the dynamics of endogenous insulin which is firstly transported to the liver for the ‘first pass’ and cleared by kidneys at the end 39 . Notably, fluorescence signals were still observed in the jejunum and ileum segments and the liver up to 8 h post administration. This was ascribed to that DCDE-mediated adhesion and accumulation of LG/insulin-40 in the mucus prolonged its retention in the gut, which thereby led to continued absorption of insulin and much longer hyperglycemic effect than s.c. insulin ( Fig. 4a, c ). We next assessed the integrity of LG/insulin Cy5.5 -40 during the intestinal absorption and liver delivery. From the cryosection imaging, most of the LG/insulin Cy5.5 -40 remained as puncta in the epithelia, while few LG/insulin Cy5.5 -40scattered in the perisinusoidal space, and strongly diffusive fluorescence was recorded spreading throughout the hepatocytes ( Fig. 4g, Supplementary Fig. 43 ), indicating that LG/insulin-40 underwent dissociation mainly in the liver and less during enteral absorption. As sinusoidal glutathione efflux is accompanied by a high local concentration of glutathione (7-10 mM) 40 , we reasoned that when the absorbed LG/insulin-40 was transited through the portal vein to hepatic sinusoids and diffused through the fenestrations to the perisinusoidal space, the poly(disulfide)s underwent glutathione-triggered reductive depolymerization to liberate insulin in the liver. This hypothesis was cross-validated by the much faster LG/insulin-40 dissociation kinetics at the hepatic glutathione pool level than those at villus epithelial and plasma glutathione levels ( Supplementary Fig. 32 ). Moreover, measurement of hepatic glycogen contents indicated that, compared to untreated and s.c. insulin-treated groups, hepatic glucose uptake in diabetic mice was significantly improved by LG/insulin-40 and became comparable to that of healthy mice over 7-day consecutive treatment ( Fig. 4h ). Therefore, this hepatic glutathione pool-triggered insulin release would aid in restoring the liver as a primary metabolic regulator of glucose metabolism and correcting the aberrant glucose metabolism linked to s.c. insulin 41,42 , reinforcing the biopotency of LG/insulin-40 in mimicking endogenous insulin. We further evaluated the long-term biosafety of this oral insulin formulation via dosing diabetic mice with LG/insulin-40-loaded capsules twice daily for two weeks. Like control diabetic mice receiving saline, all experimental mice showed no dramatic fluctuations in body weight ( Supplementary Fig. 43 ). Histological examination of intestinal segments revealed unaltered mucosal thickness, intact fingerlike villi, insignificant infiltration of immune cells and no opening of tight junctions ( Supplementary Figs. 44, 45 ). The serum levels of endotoxin and typical proinflammatory cytokines and blood biochemical parameters in the experimental group were not statistically different from those of the control group ( Supplementary Figs. 4 6-48 ). The histological sections of each major organ displayed similar appearance in both groups with no discernible differences in pathological changes ( Supplementary Fig. 49 ). Glycemic control in large animals We moved on to test the feasibility of LG/insulin-40 for perorally delivering insulin in large animals, aiming to bridge the gap between rodent models and human diabetes. As pigs have similarities with humans in dietary needs, gastrointestinal physiology, and metabolism 43 , we leveraged streptozotocin-induced diabetic Bama minipigs as the large animal model ( Fig. 5a, Supplementary Fig. 50 ). Unlike s.c. insulin (0.5 IU kg -1 ) which produced a rapid hypoglycemic onset and quick resolution, oral gavage of LG/sinulin-40-loaded enteric capsules (size 2, Supplementary Fig. 37 ; 2 IU kg -1 ) led to a gradual decline in BGLs and longer hypoglycemic effect ( Fig. 5b, Supplementary Fig. 51 ), and yielded a pharmacological availability of 19.3% ( Fig. 5c ). Meanwhile, a slower plasma insulin clearance rate was detected for capsule-delivered LG/insulin-40 compared to s.c. insulin ( Fig. 5d ), producing a relative bioavailability of 20.8% ( Fig. 5e ). Oral gavage of LG/insulin-40 solution, free insulin-loaded capsules, or blank capsules did not attenuate BGLs or induce plasma exposure of insulin. The comparable pharmacodynamic activity and relative bioavailability achieved in diabetic minipigs and mice underscore the interspecies translatability of this poly(disulfide)s-formulated oral insulin. We further assessed the efficacy of multi-day oral dosing in controlling postprandial and daily glycemia, as would be necessary for mitigating risks of diabetic cardiovascular complications 44 . Diabetic minipigs were treated with LG/insulin-40-loaded capsules (4 IU kg -1 ) 2 h before the two daily meals for two weeks ( Fig. 5f, Supplementary Fig. 52 ). Although control minipigs receiving preprandial s.c. insulin (1 IU kg -1 , injected 0.5 h before each meal) showed severely suppressed postprandial BGLs, hypoglycemic states (200 mg dl -1 ) around 3-5 h after injection. By contrast, in at least two-thirds of cases, the oral intake of LG/insulin-40 controlled the postprandial glycemia below 200 mg dl -1 and further sustain the BGL within the range of 50-100 mg dl -1 for 1-4 h after the postprandial rise. Overall, the oral insulin maintained the BGL below 200 mg dl -1 for approximately 70% of the experimental period, with only a few short-lived instances of hypoglycemia. Analysis of the mean amplitude of glucose excursions (MAGEs) revealed much smaller daily fluctuations of BGLs in the LG/insulin-40 group than those in the s.c. insulin group, evidencing that minipigs intaking oral insulin experienced less glycemic variability ( Fig. 5g, Supplementary Fig. 53 ). This was further supported by the 1.9-fold reduction in serum glycated albumin levels, an indicator reflective of the glycemic control over the preceding 2-3 weeks 45 , in the oral insulin group (from 32.0% to 23.1%) than that of the s.c. insulin group (from 31.6% to 27.1%; Fig. 5h ). The better management of glycemia attained by oral insulin could be attributed to the liver-targeted delivery paradigm, which closely recreates the physiologic response to insulin under feeding conditions and provides better control over glucose metabolism across the liver and whole body 41,42 . Throughout the two-week consecutive treatment, minipigs treated with oral insulin behaved normally and showed neither alternation in feeding or stooling patterns nor reduction in body weight ( Supplementary Fig. 54 ). At the end, histological analysis of different intestinal segments revealed normal morphology and no signs of inflammation ( Supplementary Fig. 55 ). No obvious abnormalities in the blood biochemistry, complete blood count, and histology of major organs between the oral insulin and subcutaneous insulin treated groups were detected ( Supplementary Figs. 56-58 ). Scale-up preparation and formulation generality Given the simplicity of this oral insulin formulation strategy, we further explored the feasibility of scaling up the formulation volume from 1 ml to 1 liter. LG/insulin-40 prepared in the larger batch presented similar physicochemical characteristics and attained a similar pharmacological availability in diabetic mice ( Supplementary Fig. 59 ). Of note, the disulfide monomer can be synthesized straightforwardly using inexpensive commercial chemicals and laboratory accessible instruments. As estimated, synthesis of 1 gram of LGs costs approximately 67.4 US$, which can be used to formulate about 9,100 IU of oral insulin; if a diabetic patient took this oral insulin at a dosage of 100 IU per time, the needed excipient monomers would cost about 0.7 US$ ( Supplementary Table S1 ). Therefore, this oral insulin would render effective diabetes management at an economically reasonable level. To demonstrate the adaptability of this formulation strategy, we first chose HSA (isoelectric point ≈ 5, M.W. ≈ 67 kDa), a protein widely studied as a vector for various therapeutics, 46 as a model biomacromolecule drug. After dosing HSA/poly(disulfide)s complex coacervates to 4T1 tumor-bearing mice, gradual accumulation of HSA in the tumor was observed over time ( Supplementary Figs. 60-62 ). To further showcase the flexibility in choosing lipoic acid derivatives, we then used alendronate-conjugated lipoic acid as the monomer to coacervate sCT (isoelectric point = 8.86), a model peptide showing positive charges at physiological pH. When sCT/poly(disulfide)s complex coacervates were ingested by mice, the hypocalcemia effect lasted longer compared to s.c. sCT, yielding a relative bioactivity of 22.8% ( Supplementary Figs. 63-69 ). Discussion We have developed a simple yet effective strategy for formulating oral biologics via utilizing in-situ polymerized poly(disulfide)s as the sole excipient. The systemic uptake of these biologics is mediated by the DCDE initiated by intrinsic thiols at different ‘stations’ along the bodily journey of poly(disulfide)s-coacervated biomacromolecules (summarized in Fig. 6 ). Using insulin as a model biologic drug and lipoic acid-guanidine derivative as the poly(disulfide)s monomer, we systematically studied how the complex coacervates were formulated and overcame the intestinal barriers without breaching them. While DCDE-mediated oral delivery of poly(disulfide)s-coacervated biomacromolecules is introduced for the first time, the remarkable pharmacodynamic and pharmacokinetic outcomes attained in murine and swine models, along with the successful scale-up preparation and the generality of this formulation approach, support the reliability of poly(disulfide)s-engineered oral biologics. As the rapid, dynamic nature of DCDE makes itself hard to characterize, these findings justify the need for gathering more detailed mechanical insights via advanced technologies with higher resolutions in future studies. Importantly, the dosing paradigm of poly(disulfide)s-formulated oral biologics offers the following advantages. Administration with patient-familiar capsules would afford greater convenience and medication compliance, obviating bowel discomfort or phobia of latent gastrointestinal perforation or obstruction. The ubiquitous existence of thiols in mucins and epithelial membranes could enable biomacromolecule absorption at different intestinal segments, providing prolonged systemic uptake and adequate bioavailability while avoiding limited or delayed uptake associated with specific receptor or region-targeting delivery systems. The noninvasive absorption procedure would leave the intestinal barriers intact, averting potential complications accompanying the manipulation of digestive enzyme activities, the mucus thickness, and the epithelial integrity via chemical or physical approaches. Still, local and systemic chronic effects over longer repeated dosing regimens require further study in subsequent preclinical and clinical assessments. By design, it can be naturally envisioned that this oral formulation is modular and scalable. In addition to the recombinant human insulin, insulin or insulin analogs with different therapeutic windows and half-lives should also be compatible with this poly(disulfide)s formulating strategy. Additionally, this formulation scheme could be expanded to develop a gallery of oral biologic drugs, which are currently administered by parenteral routes. From a chemical viewpoint, this is made possible by choosing appropriate lipoic acid derivatives depending on the interactions harnessed to interface the biomacromolecules. Overall, the simplicity, high bioavailability, scalability, and cost-effectiveness of poly(disulfide)s-formulated biologics would inspire the development of a new generation of oral therapeutics. Declarations Acknowledgments This work was funded by the National Key Research and Development Program of China (2020YFA0210800, H.H.Y.; 2022YFE0202200, Z.G.), the Major Project of Science and Technology of Fujian Province (2020HZ06006, H.H.Y.), the National Natural Science Foundation of China (22027805, H.H.Y.; 22334004, H.H.Y.; 22107019, Z.W.C.; 22277011, Z.W.C.), Start-up packages of Zhejiang University (Z.G.), and the Kunpeng Program grant (Z.G.). Work performed at the Center for Nanoscale Materials, a U.S. Department of Energy Office of Science User Facility, was supported by the U.S. DOE, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. Author contributions C.C., Z.W.C. and H.H.Y. designed the project. C.C., T.J.H., Z.L., H.J.L., Y.Z., X.L., Y.H.G., J.Y.L. and Z.W.C. performed the experiments. C.C., T.J.H., Z.L., H.J.L., Y.Z., X.L., Y.H.G., J.Y.L., T.W., W.H., H.C., E.A.R., G.J.C., Z.T.C., Z.W.C., Z.G. and H.H.Y. analyzed the data. C.C. and Z.W.C. drafted the original manuscript. C.C., Z.L., H.J.L., E.A.R., G.J.C., Z.W.C., Z.G. and H.H.Y. revised the manuscript. Competing interests C.C., T.J.H., Z.L, Z.W.C. and H.H.Y. are co-inventors on multiple patent applications covering aspects of the technology presented here. Z.G. is a scientific cofounder of ZCapsule Inc., Zenomics Inc., and μZen Pharma Inc. The other authors declare no competing interests. Data availability All data generated or analyzed during this study are included in this published article (and its Supplementary Information files) or are available from the authors upon request. References Abramson, A. et al. An ingestible self-orienting system for oral delivery of macromolecules. Science 363 , 611-615 (2019). Harrison, G.A. Insulin in alcoholic solution by the mouth. Br. Med. J. 2 , 1204 (1923). Sims, E.K., Carr, A.L.J., Oram, R.A., DiMeglio, L.A. & Evans-Molina, C. 100 years of insulin: celebrating the past, present and future of diabetes therapy. Nat. Med. 27 , 1154-1164 (2021). Abramson, A. et al. Oral delivery of systemic monoclonal antibodies, peptides and small molecules using gastric auto-injectors. Nat. Biotechnol. 40 , 103-109 (2022). Chen, Z. et al. Synthetic beta cells for fusion-mediated dynamic insulin secretion. Nat. Chem. Biol. 14 , 86-93 (2018). Drucker, D.J. Advances in oral peptide therapeutics. Nat. Rev. Drug. Discov. 19 , 277-289 (2020). Iyer, H., Khedkar, A. & Verma, M. Oral insulin – a review of current status. Diabetes Obes. Metab. 12 , 179-185 (2010). Brown, T.D., Whitehead, K.A. & Mitragotri, S. Materials for oral delivery of proteins and peptides. Nat. Rev. Mater. 5 , 127-148 (2020). Lee, J.S. et al. Metabolic and immunomodulatory control of type 1 diabetes via orally delivered bile-acid-polymer nanocarriers of insulin or rapamycin. Nat. Biomed. Eng. 5 , 983-997 (2021). Xiao, Y. et al. Oral Insulin Delivery Platforms: Strategies To Address the Biological Barriers. Angew. Chem. Int. Ed. 59 , 19787-19795 (2020). Anselmo, A.C., Gokarn, Y. & Mitragotri, S. Non-invasive delivery strategies for biologics. Nat. Rev. Drug. Discov. 18 , 19-40 (2019). Han, X. et al. Zwitterionic micelles efficiently deliver oral insulin without opening tight junctions. Nat. Nanotechnol. 15 , 605–614 (2020). Lamson, N.G., Berger, A., Fein, K.C. & Whitehead, K.A. Anionic nanoparticles enable the oral delivery of proteins by enhancing intestinal permeability. Nat. Biomed. Eng. 4 , 84-96 (2020). Banerjee, A. et al. Ionic liquids for oral insulin delivery. Proc. Natl. Acad. Sci. USA 115 , 7296 (2018). Pridgen Eric, M. et al. Transepithelial Transport of Fc-Targeted Nanoparticles by the Neonatal Fc Receptor for Oral Delivery. Sci. Transl. Med. 5 , 213ra167-213ra167 (2013). Xiao, Y. et al. Glucose-responsive oral insulin delivery platform for one treatment a day in diabetes. Matter 4 , 3269-3285 (2021). Hubálek, F. et al. Molecular engineering of safe and efficacious oral basal insulin. Nat. Commun. 11 , 3746 (2020). Abramson, A. et al. A luminal unfolding microneedle injector for oral delivery of macromolecules. Nat. Med. 25 , 1512-1518 (2019). Banerjee, A., Lee, J. & Mitragotri, S. Intestinal mucoadhesive devices for oral delivery of insulin. Bioeng. Transl. Med. 1 , 338-346 (2016). Bang, E.-K. et al. Substrate-Initiated Synthesis of Cell-Penetrating Poly(disulfide)s. J. Am. Chem. Soc. 135 , 2088-2091 (2013). Zhou, J. et al. Self-Assembled and Size-Controllable Oligonucleotide Nanospheres for Effective Antisense Gene Delivery through an Endocytosis-Independent Pathway. Angew. Chem. Int. Ed. 58 , 5236-5240 (2019). Gasparini, G. et al. Cellular Uptake of Substrate-Initiated Cell-Penetrating Poly(disulfide)s. J. Am. Chem. Soc. 136 , 6069-6074 (2014). Qian, L. et al. Intracellular Delivery of Native Proteins Facilitated by Cell-Penetrating Poly(disulfide)s. Angew. Chem. Int. Ed. 57 , 1532-1536 (2018). Laurent, Q. et al. Thiol-Mediated Uptake. JACS Au 1 , 710-728 (2021). Torres, A.G. & Gait, M.J. Exploiting cell surface thiols to enhance cellular uptake. Trends Biotechnol. 30 , 185-190 (2012). Cheng, Y. et al. Inhibitors of thiol-mediated uptake. Chem. Sci. 12 , 626-631 (2021). Artemova, N., Stein-Margolina, V., Smirnova, E. & Gurvits, B. Formation of supramolecular structures of a native-like protein in the presence of amphiphilic peptides: Variations in aggregate morphology. FEBS Lett. 586 , 186-190 (2012). Hubatsch, I., Ragnarsson, E.G.E. & Artursson, P. Determination of drug permeability and prediction of drug absorption in Caco-2 monolayers. Nat. Protoc. 2 , 2111-2119 (2007). Halberg, I.B. et al. Efficacy and safety of oral basal insulin versus subcutaneous insulin glargine in type 2 diabetes: a randomised, double-blind, phase 2 trial. Lancet Diabetes Endocrinol. 7 , 179-188 (2019). Di Pisa, M., Chassaing, G. & Swiecicki, J.-M. Translocation Mechanism(s) of Cell-Penetrating Peptides: Biophysical Studies Using Artificial Membrane Bilayers. Biochemistry 54 , 194-207 (2015). Gasparini, G., Bang, E.-K., Montenegro, J. & Matile, S. Cellular uptake: lessons from supramolecular organic chemistry. Chem. Commun. 51 , 10389-10402 (2015). Derivery, E., Bartolami, E., Matile, S. & Gonzalez-Gaitan, M. Efficient Delivery of Quantum Dots into the Cytosol of Cells Using Cell-Penetrating Poly(disulfide)s. J. Am. Chem. Soc. 139 , 10172-10175 (2017). Guo, M. et al. Probing the Stochastic, Motor-Driven Properties of the Cytoplasm Using Force Spectrum Microscopy. Cell 158 , 822-832 (2014). Nkabyo, Y.S., Ziegler, T.R., Gu, L.H., Watson, W.H. & Jones, D.P. Glutathione and thioredoxin redox during differentiation in human colon epithelial (Caco-2) cells. Am. J. Physiol. - Gastrointest. Liver Physiol. 283 , G1352-G1359 (2002). Pulcu, G.S. et al. Single-Molecule Kinetics of Growth and Degradation of Cell-Penetrating Poly(disulfide)s. J. Am. Chem. Soc. 141 , 12444-12447 (2019). Chuard, N., Gasparini, G., Roux, A., Sakai, N. & Matile, S. Cell-penetrating poly(disulfide)s: the dependence of activity, depolymerization kinetics and intracellular localization on their length. Org. Biomol. Chem 13 , 64-67 (2015). Gum, J.R. et al. The human MUC2 intestinal mucin has cysteine-rich subdomains located both upstream and downstream of its central repetitive region. J. Biol. Chem. 267 , 21375-21383 (1992). Bansil, R. & Turner, B.S. The biology of mucus: Composition, synthesis and organization. Adv. Drug Deliv. Rev. 124 , 3-15 (2018). Tokarz, V.L., MacDonald, P.E. & Klip, A. The cell biology of systemic insulin function. J. Cell Biol. 217 , 2273-2289 (2018). Jiang, X., Du, B. & Zheng, J. Glutathione-mediated biotransformation in the liver modulates nanoparticle transport. Nat. Nanotechnol. 14 , 874-882 (2019). Edgerton, D.S. et al. Targeting insulin to the liver corrects defects in glucose metabolism caused by peripheral insulin delivery. JCI Insight 4 , e126974 (2019). Lewis, G.F., Carpentier, A.C., Pereira, S., Hahn, M. & Giacca, A. Direct and indirect control of hepatic glucose production by insulin. Cell Metab. 33 , 709-720 (2021). Kleinert, M. et al. Animal models of obesity and diabetes mellitus. Nat. Rev. Endocrinol. 14 , 140-162 (2018). Brown, A., Reynolds, L.R. & Bruemmer, D. Intensive glycemic control and cardiovascular disease: an update. Nat. Rev. Cardiol. 7 , 369-375 (2010). Fang, M., Daya, N., Coresh, J., Christenson, R.H. & Selvin, E. Glycated Albumin for the Diagnosis of Diabetes in US Adults. Clin. Chem. 68 , 413-421 (2022). Hoogenboezem, E.N. & Duvall, C.L. Harnessing albumin as a carrier for cancer therapies. Adv. Drug Deliv. Rev. 130 , 73-89 (2018). Additional Declarations Yes there is potential Competing Interest. C.C., T.J.H., Z.L, Z.W.C. and H.H.Y. are co-inventors on multiple patent applications covering aspects of the technology presented here. Z.G. is a scientific cofounder of ZCapsule Inc., Zenomics Inc., and μZen Pharma Inc. The other authors declare no competing interests. Supplementary Files SupplementaryInformation.pdf Supplementary Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3616020","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":265022610,"identity":"e10388e9-eb68-46b0-8076-de3ea120f321","order_by":0,"name":"Huanghao Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyUlEQVRIiWNgGAWjYBACAwkwZcMPpnhI0JIm2UCqlsMkaDGX7jH8XPDrvIT8jATGB2/bGOTNCWmxnHPGWHpm320JxhkJzIZz2xgMdzYQctiNHANp3p7bdcwSCWzSvG0MCQYHCGsx/s3bc06CTSKB/TexWsykeX4ckOAB2sJMpJa0MmvehmQJCZ6HzZJzzkkYbiCsJXnzbZ4/dhLy7ckHP7wps5EnaAsDA4cBA2MbiMHYACQkCKoHAvYHDAx/iFE4CkbBKBgFIxYAAF5wPCN93WLOAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-5894-0909","institution":"Fuzhou University","correspondingAuthor":true,"prefix":"","firstName":"Huanghao","middleName":"","lastName":"Yang","suffix":""},{"id":265022611,"identity":"7cb8b0ec-b58e-40c6-b7ba-60df00481506","order_by":1,"name":"Chen Chen","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Chen","middleName":"","lastName":"Chen","suffix":""},{"id":265022612,"identity":"9317027f-cd83-412b-98d5-c016c67d5531","order_by":2,"name":"Tingjing Huang","email":"","orcid":"","institution":"Fuzhou Univeristy","correspondingAuthor":false,"prefix":"","firstName":"Tingjing","middleName":"","lastName":"Huang","suffix":""},{"id":265022613,"identity":"457d9dbc-9835-4e7e-b932-f8e2fc6d1d1e","order_by":3,"name":"Zheng Liu","email":"","orcid":"","institution":"Fuzhou Univeristy","correspondingAuthor":false,"prefix":"","firstName":"Zheng","middleName":"","lastName":"Liu","suffix":""},{"id":265022614,"identity":"75af42fe-2584-40a1-b9a8-de82a46a5ca4","order_by":4,"name":"Hongjun Li","email":"","orcid":"https://orcid.org/0000-0002-1765-8445","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Hongjun","middleName":"","lastName":"Li","suffix":""},{"id":265022615,"identity":"00f69f03-9aa6-47b7-a25a-c38ef5e72a5d","order_by":5,"name":"Yang Zhang","email":"","orcid":"","institution":"Zhenzhou University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Zhang","suffix":""},{"id":265022616,"identity":"6e808f35-8a76-404e-8807-ea57603f653d","order_by":6,"name":"Xia Liu","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Xia","middleName":"","lastName":"Liu","suffix":""},{"id":265022617,"identity":"90ef8f4d-7708-4d35-ac17-fbdca0cae7c0","order_by":7,"name":"Yuheng Guo","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Yuheng","middleName":"","lastName":"Guo","suffix":""},{"id":265022618,"identity":"1ef74b62-198b-4590-a3e2-3e3eefed179b","order_by":8,"name":"Jingying Li","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Jingying","middleName":"","lastName":"Li","suffix":""},{"id":265022619,"identity":"e8cde314-287d-4dc4-bfe1-d1d80778339a","order_by":9,"name":"Ting Wu","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Ting","middleName":"","lastName":"Wu","suffix":""},{"id":265022620,"identity":"3ee85b3c-a4f4-4aea-91d5-4d2f5b2b6d8a","order_by":10,"name":"Wen He","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Wen","middleName":"","lastName":"He","suffix":""},{"id":265022621,"identity":"a8433bab-e7c5-43a1-9960-0548904d61c3","order_by":11,"name":"Hao Chen","email":"","orcid":"","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Chen","suffix":""},{"id":265022622,"identity":"40d72a9b-bcad-408c-bbc5-8d365010cbc4","order_by":12,"name":"Elena Rozhkova","email":"","orcid":"https://orcid.org/0000-0001-8498-8228","institution":"Center for Nanoscale Materials, Argonne National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Elena","middleName":"","lastName":"Rozhkova","suffix":""},{"id":265022623,"identity":"e05cf2fa-2d64-40c0-a68c-6555f6e83f2e","order_by":13,"name":"Guojun Chen","email":"","orcid":"https://orcid.org/0000-0002-3272-2943","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Guojun","middleName":"","lastName":"Chen","suffix":""},{"id":265022624,"identity":"5d47be70-d841-40b1-b3e5-2a4c9cd5abc4","order_by":14,"name":"Zhitong Chen","email":"","orcid":"https://orcid.org/0000-0002-2957-3078","institution":"Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhitong","middleName":"","lastName":"Chen","suffix":""},{"id":265022625,"identity":"219cc547-4d9f-4919-ab2c-fc7e15d139cc","order_by":15,"name":"Zhaowei Chen","email":"","orcid":"https://orcid.org/0000-0001-9007-5513","institution":"Fuzhou University","correspondingAuthor":false,"prefix":"","firstName":"Zhaowei","middleName":"","lastName":"Chen","suffix":""},{"id":265022626,"identity":"107afaba-8583-4cee-8bb1-c67b43fcdb7a","order_by":16,"name":"Zhen Gu","email":"","orcid":"https://orcid.org/0000-0003-2947-4456","institution":"Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Zhen","middleName":"","lastName":"Gu","suffix":""}],"badges":[],"createdAt":"2023-11-15 15:55:40","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3616020/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3616020/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":52302765,"identity":"1e338762-0cb1-4863-b451-116a14878542","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":884818,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComplex coacervation of insulin and\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e in-situ\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e growing poly(disulfide)s.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, The electrostatic surface potential of recombinant human insulin viewed from different directions. Red and blue colors represent negatively and positively charged domains, respectively; \u003cem\u003ek,\u003c/em\u003e \u003cem\u003eT,\u003c/em\u003e and \u003cem\u003ee\u003c/em\u003edenote the Boltzmann constant, temperature, and magnitude of the electron charge, respectively. Images were generated by PyMOL using 2jv1 PDB file. \u003cstrong\u003eb\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eSchematic showing the formation of complex coacervates between insulin and poly(disulfide)s which are grown \u003cem\u003ein situ\u003c/em\u003efrom lipoic acid-guanidinium (LG) monomers. Once the carboxyl groups in the negatively charged patches of insulin bind guanidinium groups of LGs \u003cem\u003evia\u003c/em\u003e salt bridges, the decrease in net charges and increase in hydrophobicity of insulin cause preliminary phase separation and bring LGs into proximity; then, LGs undergo concentration-dependent, strain-release-promoted ring-opening polymerization. The multivalent interactions between insulin and poly(disulfide)s give rise to the formation of multivalent interaction-stabilized complex coacervates. \u003cstrong\u003ec\u003c/strong\u003e, Cryogenic electron microscopy images of the complex coacervates formed from LGs and insulin at different starting molar ratios (LG/insulin-\u003cem\u003ex\u003c/em\u003e, \u003cem\u003ex\u003c/em\u003e = 10, 20, 40, 60, and 80). Scale bars, 100 nm. \u003cstrong\u003ed\u003c/strong\u003e, The insulin encapsulation efficiency (EE) and loading capacity (LC) of LG/insulin-\u003cem\u003ex\u003c/em\u003e. Data points represent mean ± s.d. (n = 3 independent experiments).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/260a97c1c16d4035ea14561b.png"},{"id":52302768,"identity":"a6d94297-d7a1-44bf-9568-a3534571cea0","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2393367,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e transepithelial transport of LG/insulin-\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ex\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e (\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ex\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e = 10, 20, 40).\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, \u003cstrong\u003eb\u003c/strong\u003e, Cumulative transport profiles (\u003cstrong\u003ea\u003c/strong\u003e) and apparent permeability coefficients (\u003cem\u003eP\u003c/em\u003e\u003csub\u003eapp\u003c/sub\u003e; \u003cstrong\u003eb\u003c/strong\u003e) of different insulin formulations including LG/insulin-\u003cem\u003ex\u003c/em\u003e, free insulin, and sodium caprate (C10) plus free insulin across the Caco-2 monolayers. Data points represent mean ± s.d. (n = 3 independent experiments). Statistical significance was calculated \u003cem\u003evia\u003c/em\u003e one-way ANOVA analysis with a Tukey post-hoc test. ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001, ****\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001. \u003cstrong\u003ec\u003c/strong\u003e, Heatmap showing the \u003cem\u003eP\u003c/em\u003e\u003csub\u003eapp\u003c/sub\u003e values of LG/insulin-\u003cem\u003ex\u003c/em\u003e across Caco-2 monolayers in the presence of endocytosis inhibitors (chlorpromazine (CPZ), methyl-β-cyclodextrin (mβCD), and wortmannin (wort)), exocytosis inhibitors (Exo1 and endosidin 2 (ES2)), DTNB (blocking exofacial thiols), DTNB-loaded fuosogenic liposomes (D-lipo; blocking intracellular thiols), and blank liposomes (B-lipo). Heatmap grids correspond to the three independent experiments in each group. \u003cstrong\u003ed\u003c/strong\u003e, Confocal \u003cem\u003ex\u003c/em\u003e-\u003cem\u003ez\u003c/em\u003e sections of Caco-2 cells after incubated with Cy5.5-tagged LG/insulin-\u003cem\u003ex\u003c/em\u003e (pink) for 10 min. The plasma membrane was stained with MemGlow™ NR12A to show the local lipid organization. Ratiometric images were generated by dividing the intensity of the 550-600 nm channel (I\u003csub\u003e550−600\u003c/sub\u003e; green) by that of the channel (\u003cem\u003eI\u003c/em\u003e\u003csub\u003e600−650\u003c/sub\u003e; red). Compared to non-binding areas, the blue shift in emission of NR12A in regions binding LG/insulin-\u003cem\u003ex\u003c/em\u003e (indicated by white arrows) showed a local increase in lipid orders. Scale bar for all panels, 20 μm. \u003cstrong\u003ee\u003c/strong\u003e, A working hypothesis for DCDE-mediated translocation of LG/insulin-\u003cem\u003ex\u003c/em\u003e across the plasma membrane. After LG/insulin-\u003cem\u003ex\u003c/em\u003e first binds to the plasma membrane by exchanging disulfides with the exofacial thiols (\u003cstrong\u003ei\u003c/strong\u003e), dynamic covalent exchange then continues with the sulfur networks of transmembrane proteins (\u003cstrong\u003eii\u003c/strong\u003e), and the recovery of temporarily denatured proteins liberates LG/insulin-\u003cem\u003ex\u003c/em\u003e into the cytosol (\u003cstrong\u003eiv\u003c/strong\u003e). Accompanying the exchange cascade, the region binding LG/insulin-\u003cem\u003ex\u003c/em\u003e undergo adaptive membrane reorganizations (\u003cstrong\u003eiii\u003c/strong\u003e), like forming sealed transient micellar pores\u003csup\u003e24\u003c/sup\u003e, to facilitate the transmembrane translocation. \u003cstrong\u003ef\u003c/strong\u003e, Confocal\u003cem\u003e x\u003c/em\u003e-\u003cem\u003ez\u003c/em\u003e sections showing the translocation of FITC-tagged LG/insulin-\u003cem\u003ex\u003c/em\u003e (green) across the basal plasma membrane of Caco-2 monolayers in the absence and presence of exocytosis inhibitors (Exo1, ES2) or intracellular delivered DTNB (D-lipo). The nuclei and plasma membrane were stained with Hoechst 33342 (blue) and Alexa Fluor-555 Wheat Germ Agglutinin (red), respectively. Scale bar for all panels, 20 μm.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/2196dbc9200ee771773dfd99.png"},{"id":52302769,"identity":"c035083a-e7fc-47d5-8c15-bdc78ece3bfb","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":3276118,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTransmucosal delivery of insulin enabled by LG/insulin-40.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, Relative blood glucose changes in diabetic mice after intrajejunal injection of LG/insulin-40 (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), free insulin (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), sodium caprate (C10) plus insulin (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), and PBS buffer. Subcutaneous (s.c.) injection of insulin (5 IU kg\u003csup\u003e-1\u003c/sup\u003e) was used as a control. \u003cstrong\u003eb\u003c/strong\u003e, AACs from \u003cstrong\u003ea\u003c/strong\u003e showing the pharmacodynamic performance of LG/insulin-40. Data points represent mean ± s.d. (n = 5 mice per group). \u003cstrong\u003ec\u003c/strong\u003e, Representative TEM images of the epithelial tissues 2 h post intrajejunal injection of LG/insulin-40 or sodium caprate plus insulin. The tissue section from untreated mice was used as the control. Tight junctions were indicated by arrows. Scar bars, 200 nm. \u003cstrong\u003ed\u003c/strong\u003e, Percentage of insulin remained after incubation of LG/insulin-40 or free insulin in simulated intestinal fluid. Data points represent mean ± s.d. (n = 3 independent experiments). \u003cstrong\u003ee\u003c/strong\u003e, 3D confocal microscopy images (size: 640 μm × 640 μm) showing the migration of LG/insulin-40 and insulin in reconstituted mucus and LG/insulin-40 in reconstituted mucus pretreated with DTNB. Insulin was tagged with FITC (green), and mucus was stained by Alexa Fluor-555 Wheat Germ Agglutinin (red). \u003cstrong\u003ef\u003c/strong\u003e, Cumulative insulin transport profiles for LG/insulin-40 and free insulin across the reconstituted mucus and LG/insulin-40 across the DTNB-pretreated mucus (LG/insulin-40 + DTNB). \u003cstrong\u003eg\u003c/strong\u003e, \u003cem\u003eP\u003c/em\u003e\u003csub\u003eapp\u003c/sub\u003e values of LG/insulin-40 and insulin across mucus and LG/insulin-40 across DTNB-pretreated mucus. Data points represent mean ± s.d. (n = 3 independent experiments). \u003cstrong\u003eh\u003c/strong\u003e, A proposed model of DCDE-mediated translocation of LG/insulin-40 in the mucus. Initiated by thiols in the cysteine-rich domains of mucins, dynamic covalent exchange cascades occur between the disulfides in LG/insulin-40 and the sulfurs of mucins, which temporarily capture LG/insulin-40 to promote its accumulation in the mucus. This dynamic process generates concentration gradients between the upper and lower mucus layers, thus facilitating the directional diffusion of LG/insulin-40 towards the epithelia. Statistical significance was calculated \u003cem\u003evia\u003c/em\u003e one-way ANOVA analysis with a Tukey post-hoc test. **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001, ****\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001. n.s., not significant.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/fdba93390b941e36f95dde9d.png"},{"id":52302771,"identity":"6488985a-306a-4189-9814-add50f948675","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":3049362,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOral insulin delivery with LG/insulin-40-loaded enteric capsules (size M) in diabetic mice\u003c/strong\u003e. \u003cstrong\u003ea\u003c/strong\u003e, Time course of BGLs of mice treated with LG/insulin-40-loaded capsules (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), free insulin-loaded capsules (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), LG/insulin-40 solution (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), blank capsules or s.c. insulin (5 IU kg\u003csup\u003e-1\u003c/sup\u003e). \u003cstrong\u003eb\u003c/strong\u003e, AACs derived from \u003cstrong\u003ea\u003c/strong\u003e showing the pharmacodynamic performance of insulin of different formulations. \u003cstrong\u003ec\u003c/strong\u003e, Changes in plasma human insulin concentrations of diabetic mice over time after receiving the treatments shown in \u003cstrong\u003ea\u003c/strong\u003e. \u003cstrong\u003ed\u003c/strong\u003e, Area under the curves (AUCs) derived from \u003cstrong\u003ec\u003c/strong\u003e showing the relative oral bioavailability of insulin of different formulations. \u003cstrong\u003ee\u003c/strong\u003e, Representative \u003cem\u003eex vivo\u003c/em\u003e images of the intestinal tracts of diabetic mice at different time points post oral gavage of LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40 loaded capsules and cryosections of the lighted segments (duodenum, 1 h; jejunum, 2 h; ileum, 4 h; colon, 6 h) showing the passage and absorption of LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40 in the intestine. Red: Cy5.5; blue: Hoechst 33342. Scale bars, 200 μm. p, photons; sr, steradian. \u003cstrong\u003ef\u003c/strong\u003e, Representative \u003cem\u003eex vivo\u003c/em\u003e imaging of major organs at different time points post oral gavage of LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40 loaded enteric capsules. \u003cstrong\u003eg\u003c/strong\u003e, Representative zoomed cryosection images of duodenum (1 h after dosing), jejunum (2 h after dosing), ileum (4 h after dosing), colon (6 h after dosing) and liver (4 h after dosing) showing the integrity of LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40 in different intestinal segments and the liver. Red: Cy5.5; blue: Hoechst 33342. Scale bars, 20 μm. \u003cstrong\u003eh\u003c/strong\u003e, Hepatic glycogen contents of diabetic mice post treated with oral insulin (20 IU kg\u003csup\u003e-1\u003c/sup\u003e) and s.c. insulin (5 IU kg\u003csup\u003e-1\u003c/sup\u003e) for different days. Diabetic mice without insulin treatment and healthy mice were used as controls. Data points represent mean ± s.d. (n = 5 mice per group). Statistical significance was calculated \u003cem\u003evia\u003c/em\u003e one-way ANOVA analysis with a Tukey post-hoc test. *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001, ****\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001. n.s., not significant.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/b26d7d77755e571e80d16614.png"},{"id":52302766,"identity":"689527e4-d104-48ca-862a-21b13c6a09e2","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2579026,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOral insulin delivery with LG/insulin-40-loaded enteric capsules in diabetic Bama minipigs.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, Schematic showing the oral gavage of LG/insulin-40-loaded enteric capsules into a minipig and recording its BGLs by a continuous glucose monitoring system. \u003cstrong\u003eb\u003c/strong\u003e, Profiles of the relative BGLs of minipigs treated with LG/insulin-40-loaded capsules (2 IU kg\u003csup\u003e-1\u003c/sup\u003e), free insulin-loaded capsules (2 IU kg\u003csup\u003e-1\u003c/sup\u003e), LG/insulin-40 solutions (2 IU kg\u003csup\u003e-1\u003c/sup\u003e), blank capsules, or s.c. insulin (0.5 IU kg\u003csup\u003e-1\u003c/sup\u003e). \u003cstrong\u003ec\u003c/strong\u003e, AACs of blood glucose after receiving the treatments in \u003cstrong\u003eb\u003c/strong\u003e. \u003cstrong\u003ed\u003c/strong\u003e, Variation in plasma insulin concentrations of minipigs over time after receiving the treatments in \u003cstrong\u003eb\u003c/strong\u003e. \u003cstrong\u003ee\u003c/strong\u003e, AUCs of plasma insulin after receiving the treatments in \u003cstrong\u003eb\u003c/strong\u003e. \u003cstrong\u003ef\u003c/strong\u003e, Blood glucose profiles of the individual minipigs (solid lines) over the two-week consecutive treatment with s.c. insulin and oral insulin. The red and green shaded areas indicate the average value of three minipigs ± s.d. for the s.c. insulin and oral insulin group, respectively. Green and red arrows indicate the time points for administration of oral insulin (4 IU kg\u003csup\u003e-1\u003c/sup\u003e) and s.c. insulin (1 IU kg\u003csup\u003e-1\u003c/sup\u003e), respectively; blue arrows indicate the time points for feeding. \u003cstrong\u003eg\u003c/strong\u003e, MAGEs of minipigs dosed with oral or s.c. insulin on day 1, 7, and 14. MAGEs of minipigs during the whole course are shown in Supplementary Fig. 54. \u003cstrong\u003eh\u003c/strong\u003e, Percentages of serum glycated albumin (glycated albumin/total albumin) of diabetes minipigs before (day 0) and after (day 15) dosed with oral insulin or s.c. insulin twice daily for 2 weeks. For \u003cstrong\u003eb-e\u003c/strong\u003e, \u003cstrong\u003eg\u003c/strong\u003e,\u003cstrong\u003e h\u003c/strong\u003e, Data points represent mean ± s.d. (n = 3 minipigs per group). Statistical significance was calculated \u003cem\u003evia\u003c/em\u003e ANOVA analysis with a Tukey post-hoc test (\u003cstrong\u003ec\u003c/strong\u003e, \u003cstrong\u003ee\u003c/strong\u003e) or two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test (\u003cstrong\u003eg\u003c/strong\u003e, \u003cstrong\u003eh\u003c/strong\u003e). *\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05, **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001. n.s., not significant.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/365d4be015af32d2417c9e3e.png"},{"id":52302767,"identity":"46e04256-abf2-4d86-a421-54c0d5b4a4cd","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1708350,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA proposed mechanism of DCDE-mediated oral delivery of poly(disulfide)s-coacervated biomacromolecules. \u003c/strong\u003eThe complex coacervates formed between biomacromolecules and \u003cem\u003ein-situ\u003c/em\u003e growing poly(disulfide)s are packaged into enteric capsules for oral dosing. \u003cstrong\u003ea\u003c/strong\u003e-\u003cstrong\u003ec\u003c/strong\u003e, Enteral absorption of poly(disulfide)s-coacervated biomacromolecules. \u003cstrong\u003ea\u003c/strong\u003e, Upon entering the intestine, the complex coacervates are released from the enteric capsules, where the entrapment of biomacromolecules inside complex coacervates shields them against degradation by intestinal digestive enzymes. \u003cstrong\u003eb\u003c/strong\u003e, The cysteine-rich domains of mucins provide footholds for transiently capture the complex coacervates, which facilitates their adhesion to and accumulation in the mucus and prolongs their retention in the gut. The concentration gradients generated between the upper and lower mucus layers promote the migration of complex coacervates towards the epithelia underneath the mucus lining. \u003cstrong\u003ec\u003c/strong\u003e, Exofacial thiols on the apical membrane and endofacial thiols on the basal membrane of enterocytes initiate the dynamic covalent disulfide-exchange cascades occurring between the poly(disulfide)s and the sulfur networks in transmembrane proteins, mediating the transepithelial absorption of the complex coacervates into the bloodstream. \u003cstrong\u003ed\u003c/strong\u003e, When the complex coacervates are transited to the liver \u003cem\u003evia\u003c/em\u003ethe portal vein circulation, the reductive depolymerization of poly(disulfide)s triggered by the high-level glutathione (7–10 mM) in sinusoids leads to their dissociation and the concomitant cargo release. A portion of the liberated biologic drugs is subject to the first-pass metabolism in the liver, while the rest exits the liver through the hepatic vein and is delivered into the systemic circulation. \u003cstrong\u003ee\u003c/strong\u003e,\u003cstrong\u003ef\u003c/strong\u003e, In the case of insulin, the portion present in the liver acts to suppress gluconeogenesis and promote hepatic glycogenesis (\u003cstrong\u003ee\u003c/strong\u003e), and the faction that enters the systemic circulation exerts its action in the peripheral tissues, such as muscle and adipose tissue, and the brain (\u003cstrong\u003ef\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/f1cf4b778cf4e2a51a527c41.png"},{"id":54927168,"identity":"e5290f48-c558-4adf-96dc-7e54a2b205f8","added_by":"auto","created_at":"2024-04-18 17:18:48","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5354239,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/0e906c5b-5a93-4b0d-9457-26ddf278511d.pdf"},{"id":52302772,"identity":"0c34b7cd-6d79-4f36-9509-de1d55315b78","added_by":"auto","created_at":"2024-03-08 18:48:22","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":25410424,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3616020/v1/5af64ac753a724d3f44433ce.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nC.C., T.J.H., Z.L, Z.W.C. and H.H.Y. are co-inventors on multiple patent applications covering aspects of the technology presented here. Z.G. is a scientific cofounder of ZCapsule Inc., Zenomics Inc., and μZen Pharma Inc. The other authors declare no competing interests.","formattedTitle":"Dynamic Covalent Disulfide Exchange Mediates Oral Delivery of Biomacromolecules","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOral biologics have been pursued since the introduction of insulin injection for diabetes mellitus management 100 years ago.\u003csup\u003e1-3\u003c/sup\u003e Compared to injection, oral administration is often preferred due to its convenience, painlessness, and superior adherence.\u003csup\u003e4-6\u003c/sup\u003e For insulin in particular, oral delivery could closely recapitulate the physiological route of endogenous insulin delivery from the pancreas to the liver \u003cem\u003evia\u003c/em\u003e the hepatic portal\u0026nbsp;circulation, obviate peripheral hyperinsulinemia, and allow for earlier initiation of insulin therapy.\u003csup\u003e6,7\u003c/sup\u003e However, the implementation of oral biologics has been constrained by the inherent physiological function and anatomical structure of the gastrointestinal tract. While the acidic milieu in the stomach has been circumvented by technologies like enteric coatings, the formidable barriers in the intestine, including digestive enzymes, the mucus lining, and the epithelial layer, fundamentally impede the absorption of undegraded biomacromolecules.\u003csup\u003e8-10\u003c/sup\u003e Substantial efforts have been attempted to navigate these hurdles \u003cem\u003evia\u003c/em\u003e co-formulating biologic drugs with multiple ingredients (including vehicles, enzyme inhibitors, mucus-penetrating agents, and permeation enhancers) or loading biomacromolecules into swallowable miniature devices.\u003csup\u003e11-19\u003c/sup\u003e Nonetheless, these innovations inevitably increase formulation complexity and raise biological barrier manipulation associated biosafety concerns such as bowel infections, perforation, and obstruction.\u003csup\u003e6,8\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eWe here describe a facile oral biologic formulation which utilizes \u003cem\u003ein-situ\u003c/em\u003e growing\u0026nbsp;poly(disulfide)s as the\u0026nbsp;sole excipient\u0026nbsp;but can surmount all intestinal barriers without compromising their integrities.\u0026nbsp;Poly(disulfide)s, typically grown from dithiolanes such as lipoic acid derivatives, have shown to mediate the translocation of cargoes across the cell membrane into the cytosol.\u003csup\u003e20-23\u003c/sup\u003e The crux of this cellular entry paradigm is the dynamic covalent disulfide exchange (DCDE) chemistry which begins with exofacial thiols and continues along sulfur tracks in transmembrane proteins and can be inhibited by thiol-blocking reagents.\u003csup\u003e24-26\u003c/sup\u003e DCDE is also essential in biology\u0026nbsp;for the maintenance of cellular redox homeostasis, protein folding, and virus entry.\u003csup\u003e24-26\u003c/sup\u003e The magic of this natural dynamic covalent chemistry, especially for mediating systemic delivery of cargoes, is far from the limit. In this study, focusing on insulin coacervated with poly(disulfide)s grown \u003cem\u003ein situ\u003c/em\u003e from lipoic acid-guanidinium derivatives (LGs; \u003cstrong\u003eFig. 1\u003c/strong\u003e), we set out to establish the breadth and mechanism of DCDE in facilitating oral delivery of biomacromolecules, and to evaluate the pharmacodynamics, pharmacokinetics,\u0026nbsp;glycemic controllability, and biosafety of this oral insulin in diabetic mice and swine models. Additionally, to showcase the modularity of this formulation strategy, another two oral biologics, human serum albumin (HSA) and salmon calcitonin (sCT), were confected, and tested in murine models.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eComplex coacervation of insulin with \u003cem\u003ein-situ\u003c/em\u003e growing poly(disulfide)s\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven that dithiolanes tend to undergo\u0026nbsp;concentration-dependent, strain-release-promoted ring-opening polymerization\u003csup\u003e20,21\u003c/sup\u003e,\u0026nbsp;we surmise that, when the negatively charged patches on insulin surface bind the positively charged heads of amphiphilic LGs\u0026nbsp;\u003cem\u003evia\u003c/em\u003e salt bridges, the resulting decrease in net surface charge together with an increase in hydrophobicity will induce preliminary phase separation\u003csup\u003e27\u003c/sup\u003e, whereby the locally concentrated dithiolanes spontaneously grow into poly(disulfide)s. The latter concomitantly form multivalent interaction-stabilized complex coacervates with insulin (\u003cstrong\u003eFig. 1a, b, Supplementary Fig. 1\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo test this hypothesis, we mixed LGs (\u003cstrong\u003eSupplementary Figs. 2-8\u003c/strong\u003e)\u0026nbsp;and insulin at different molar ratios (LG/insulin = \u003cem\u003ex\u003c/em\u003e)\u0026nbsp;and observed that all solutions instantly turned turbid (\u003cstrong\u003eSupplementary Fig. 9\u003c/strong\u003e). Meanwhile, the characteristic absorbance of dithiolanes at λ = 330 nm gradually declined and reached a plateau within 15 min (\u003cstrong\u003eSupplementary Fig. 10\u003c/strong\u003e), while peaks of oligomeric poly(disulfide)s corresponding to n = 2–9 were revealed by MALDI-TOF analysis (\u003cstrong\u003eSupplementary Fig. 11\u003c/strong\u003e), validating the occurrence of ring-opening polymerization after preliminary condensation. Cryogenic electron microscopy and dynamic light scattering showed that nanosized complex coacervates were formed (\u003cstrong\u003eFig. 1c, Supplementary Fig. 12\u003c/strong\u003e).\u0026nbsp;In contrast, in control experiments performed at pH 4.0 (below the isoelectric point of insulin) or when using insulin plus guanidinium chloride, insulin plus lipoic acids, or the individual ingredients, neither ring opening nor complex coacervation occurred (\u003cstrong\u003eSupplementary Figs. 9, 10, 12\u003c/strong\u003e), further validating our formulation scheme. As \u003cem\u003ex\u003c/em\u003e increased from 10 to 80, the average size, surface charge, and encapsulation efficiency of complex coacervates (termed LG/insulin-\u003cem\u003ex\u003c/em\u003e) initially increased and then plateaued at \u003cem\u003ex\u003c/em\u003e above 40, while the loading capacities showed the opposite trend (\u003cstrong\u003eFig. 1d, Supplementary Figs. 12, 13\u003c/strong\u003e). Moreover, the secondary structure of insulin was preserved (\u003cstrong\u003eSupplementary Fig. 14\u003c/strong\u003e), indicating that the conserved disulfide bridges of insulin were not disturbed by the ring-opening polymerization.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDCDE-mediated transepithelial transport\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs the epithelial layer is regarded as the rate-limiting barrier for drug absorption, we next assessed the capacity of LG/insulin-\u003cem\u003ex\u003c/em\u003e to traverse Caco-2 monolayers (\u003cstrong\u003eSupplementary Fig. 15\u003c/strong\u003e),\u0026nbsp;a common \u003cem\u003ein vitro\u003c/em\u003e model for the intestinal epithelium.\u003csup\u003e28\u003c/sup\u003e In this section, studies were continued with LG/insulin-10, -20, and -40. LG/insulin-40 achieved the fastest transepithelial delivery, reaching 48.4% within 6 h, and yieldedthe highest apparent permeability coefficient (\u003cem\u003eP\u003c/em\u003e\u003csub\u003eapp\u003c/sub\u003e) of 17.8×10\u003csup\u003e-6\u003c/sup\u003e cm s\u003csup\u003e-1\u003c/sup\u003e, which is approximately 50.6, 4.6 and 1.9 times greater than with free insulin, LG/insulin-10, and LG/insulin-20, respectively\u0026nbsp;(\u003cstrong\u003eFig. 2a, b\u003c/strong\u003e).\u0026nbsp;Although sodium caprate (an intestinal permeation enhancer used in clinical trials\u003csup\u003e29\u003c/sup\u003e)\u0026nbsp;produced a similar insulin transport profile to LG/insulin-40, the integrity of the tight junctions of Caco-2 monolayers was severely compromised, which was not the case with LG/insulin-\u003cem\u003ex\u003c/em\u003e (\u003cstrong\u003eSupplementary Figs.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e16, 17\u003c/strong\u003e). Meanwhile, MTT\u0026nbsp;and lactate dehydrogenase release assays indicated that LG/insulin-\u003cem\u003ex\u003c/em\u003e exerted trifling influence on the cell viability or membrane permeability\u0026nbsp;(\u003cstrong\u003eSupplementary Fig\u003c/strong\u003e\u003cstrong\u003e. 18\u003c/strong\u003e). Together, these results suggest that the transepithelial transport of LG/insulin-x occurs through transcellular pathways.\u003c/p\u003e\n\u003cp\u003eTo gain more in-depth insights into the mechanism, we first studied the effect of classical inhibitors of endocytosis and exocytosis. These inhibitors slightly attenuated the permeation of LG/insulin-\u003cem\u003ex\u003c/em\u003e (\u003cstrong\u003eFig. 2c, Supplementary Fig\u003c/strong\u003e\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e19\u003c/strong\u003e), suggesting that regular transcytosis pathways marginally mediated this transcellular progression. However, extracellularly added or fusogenic liposome-delivered thiol-blocking agent (5,5ʹ-dithio-bis-(2-nitrobenzoic acid), DTNB; \u003cstrong\u003eSupplementary Fig\u003c/strong\u003e\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e20\u003c/strong\u003e) dramatically slowed the passage of LG/insulin-\u003cem\u003ex\u003c/em\u003e, implying that the DCDEC between the exofacial/intracellular thiols and the disulfides in LG/insulin-\u003cem\u003ex\u003c/em\u003e was decisive for mediating this underexplored transcellular transport.\u003c/p\u003e\n\u003cp\u003eWe then tracked the cellular uptake and subcellar localization of LG/insulin-\u003cem\u003ex\u003c/em\u003e over time. Following incubation for 10 min, we saw the insertion of LG/insulin-\u003cem\u003ex\u003c/em\u003e in the plasma membrane and this binding was curtailed by DTNB but not endocytosis inhibitors (\u003cstrong\u003eFig. 2d, Supplementary Figs. 21-24\u003c/strong\u003e). Moreover, a local increase in lipid order (\u003cstrong\u003eFig. 2d\u003c/strong\u003e) and reversible flip-flop of lipids (\u003cstrong\u003eSupplementary Fig. 25\u003c/strong\u003e) were observed in the binding region. Meanwhile, the non-influx of the cell impermeant propidium iodide suggested that the membrane remained intact throughout. As such adaptive membrane structural changes are reminiscent of those observed in the formation of transient micellar pores by cell-penetrating peptides when they mediate the uptake of cargoes, the translocation of LG/insulin-\u003cem\u003ex\u003c/em\u003e might follow a similar mechanism except that here the DCDE between the disulfides of LG/insulin-\u003cem\u003ex\u003c/em\u003e and transmembrane sulfur networks facilitated this process (\u003cstrong\u003eFig. 2e\u003c/strong\u003e).\u003csup\u003e24,30,31\u003c/sup\u003e Over time, LG/insulin-\u003cem\u003ex\u003c/em\u003e was primarily delivered to the cytosol and its internalization was dramatically reduced by extracellularly added DTNB, but was insensitive to endocytosis inhibitors (\u003cstrong\u003eSupplementary Figs. 26-30\u003c/strong\u003e). Analysis of the correlations between LG/insulin-\u003cem\u003ex\u003c/em\u003e and organelles, including endo-lysosomes, Golgi apparatus, endoplasmic reticulum, mitochondria, and microtubules, revealed no specific colocalizations. The cytosolic delivery paradigm of LG/insulin-\u003cem\u003ex\u003c/em\u003e is typical for DCDE-mediated cellular uptake\u003csup\u003e21-23\u003c/sup\u003e, which would help to avoid degradation associated with the endo-lysosomal endocytosis pathway. As the subcellular distribution pattern was not affected by endo-/exocytosis inhibitors, the movement of LG/insulin-\u003cem\u003ex\u003c/em\u003e across the cytosol to the basal membrane was less likely mediated by transcytosis machineries, but perhaps driven by thermal motions or aggregate cytoplasmic forces which were suggested to exert on cytosolic delivered cargoes\u003csup\u003e32,33\u003c/sup\u003e. Further, confocal \u003cem\u003ex-z\u003c/em\u003e sections revealed that the basal membranes was stained by the intracellularly delivered LG/insulin-\u003cem\u003ex\u003c/em\u003e, as observed during early uptake, and this binding was affected by intracellularly delivered DTNB but not exocytosis inhibitors (\u003cstrong\u003eFig. 2f, Supplementary Fig. 31\u003c/strong\u003e). These findings suggest that the egression of LG/insulin-\u003cem\u003ex\u003c/em\u003e across the basal membrane was a similar but inverse process of thiol-mediated uptake, relying on the DCDE initiated by endofacial thiols.\u003c/p\u003e\n\u003cp\u003eAdditionally, we noted that most LG/insulin-40 remained as discrete puncta in the cytosol, while diffuse fluorescence spreading throughout the cytoplasm was observed for LG/insulin-10 and -20 over time\u0026nbsp;(\u003cstrong\u003eSupplementary Figs. 26-30\u003c/strong\u003e). Another finding was that intracellularly delivered DTNB reduced this diffuse fluorescence to a certain extent. These phenomena hinted at the sensitivity of LG/insulin-\u003cem\u003ex\u003c/em\u003e to the cytosolic glutathione, which could trigger reductive depolymerization of poly(disulfide)s, leading to the dissociation of complex coacervates and insulin leakage. To further understand the mechanism, we studied the dissociation kinetics of LG/insulin-\u003cem\u003ex\u003c/em\u003e in solutions containing different concentrations of glutathione. Indeed, LG/insulin-40 disassembled much slower than the other two formulations at the cytosolic glutathione level of 0.3–1.6 mM in Caco-2 cells\u003csup\u003e34\u003c/sup\u003e (\u003cstrong\u003eSupplementary Fig. 32\u003c/strong\u003e). The higher stability of LG/insulin-40 stemmed from the higher content and longer chain length of poly(disulfide)s (\u003cstrong\u003eFig. 1d, Supplementary Fig. 11\u003c/strong\u003e), which provided greater cross-linkages with insulin to conceal themselves better against glutathione cleavage\u003csup\u003e35,36\u003c/sup\u003e, thereby imparting superior transcellular delivery efficacy. Based on these preliminary experiments, LG/insulin-40 was leveraged for further studies, presented below.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStability against digestion and transmucosal delivery\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe next examined the potency of LG/insulin-40 (20 IU kg\u003csup\u003e-1\u003c/sup\u003e) for transmucosal delivery \u003cem\u003ein vivo\u003c/em\u003e by intrajejunal injection to diabetic mice,\u0026nbsp;which was readily assessed by monitoring changes in blood glucose levels (BGLs). After intrajejunal injection, BGLs steadily decreased and reached a maximum reduction\u0026nbsp;of approximately 70% within 2 h (\u003cstrong\u003eFig. 3a, Supplementary Fig. 33\u003c/strong\u003e). BGLs remained below 200 mg\u0026nbsp;dl\u003csup\u003e-1\u003c/sup\u003e for nearly 7 hours, which is about 2.5-hour longer than with subcutaneous (s.c.) insulin (5 IU kg\u003csup\u003e-1\u003c/sup\u003e). Integration of the areas above the BGL curves (AACs) yielded a pharmacological availability of 27.5% (\u003cstrong\u003eFig. 3b\u003c/strong\u003e). However, intrajejunal injection of free insulin did not affect BGLs. Although intrajejunal injection of sodium caprate plus insulin produced certain BGL-lowering response, the relative bioactivity was about 2.3-times lower than that of LG/insulin-40. Importantly, unlike\u0026nbsp;sodium caprate which caused severe intestinal tight junction opening, LG/insulin-40 had no influence on the intestinal tissue morphology which remained intact like in the untreated control\u0026nbsp;(\u003cstrong\u003eFig. 3c\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eSuch a high bioactivity produced by this simple formulation\u0026nbsp;is surprising given the presence of two other physiological barriers in the intestinal tract compared to the Caco-2 monolayer model, \u003cem\u003ei.e.\u003c/em\u003e, digestive enzymes and mucus. The stability of LG/insulin-40 against proteolytic degradation was studied in simulated intestinal fluid. Compared to free insulin, which was fully degraded within 1.5 h, approximately 80% of insulin in LG/insulin-40 was preserved over 6 h incubation and the suspension of LG/insulin-40 remained\u0026nbsp;turbid still, indicating that the complexation of insulin and poly(disulfide)s into stable coacervates protected insulin against proteolytic degradation in the intestine\u0026nbsp;(\u003cstrong\u003eFig. 3d, Supplementary Fig. 34\u003c/strong\u003e). We then sought to answer how LG/insulin-40 migrated through the mucus by using a 3D migration model reconstituted with Type II mucin. We found that LG/insulin-40 homogeneously adhered to the mucus, travelled at least 200 μm within 15 min, and achieved 57.4% cumulative transport over 6 h with a \u003cem\u003eP\u003c/em\u003e\u003csub\u003eapp\u003c/sub\u003e value of 24.9×10\u003csup\u003e-6\u003c/sup\u003e cm s\u003csup\u003e-1\u003c/sup\u003e (\u003cstrong\u003eFigs. 3e-g\u003c/strong\u003e). However, in experiments conducted with DTNB-pretreated mucus, most of LG/insulin-40 spread beyond the area where it was dropped, and the adhered LG/insulin-40 showed a shallower penetration depth.\u0026nbsp;Additionally, free insulin exhibited limited binding to and penetration into the mucus.\u0026nbsp;The pronounced mucus adhesion and permeation capacity of LG/insulin-40 was unlikely achieved \u003cem\u003evia\u003c/em\u003e mucus thinning or lysis, since there was neither drop in mucus viscosity nor cleavage of mucins\u0026nbsp;(\u003cstrong\u003eSupplementary Figs. 35, 36\u003c/strong\u003e).\u0026nbsp;Given the cysteine-rich nature of mucins\u003csup\u003e37,38\u003c/sup\u003e,\u0026nbsp;it is reasonable to deduce that their sulfurs can provide dynamic covalent exchange networks to transiently capture LG/insulin-40 and promote its accumulation in the mucus, which generates concentration gradients between the upper and lower mucus layers to drive the diffusion of LG/insulin-40 towards the epithelia\u0026nbsp;(\u003cstrong\u003eFig. 3h\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOral insulin delivery in diabetic mice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe next assessed the capacity of LG/insulin-40 enabling oral delivery by gavage of diabetic mice using concentrate-loaded enteric capsules (size M, 20 IU kg\u003csup\u003e-1\u003c/sup\u003e; \u003cstrong\u003eSupplementary Fig. 37\u003c/strong\u003e). Oral ingestion of LG/insulin-40 solution (20 IU kg\u003csup\u003e-1\u003c/sup\u003e), free insulin-loaded capsules (20 IU kg\u003csup\u003e-1\u003c/sup\u003e) or blank capsules, and s.c. insulin (5 IU kg\u003csup\u003e-1\u003c/sup\u003e) were adopted as controls. Unlike s.c. insulin, which induced a sharp drop but a quick recovery in BGLs, oral administration of LG/insulin-40-loaded capsules provoked much longer hypoglycemic effect with a relative bioactivity of 26.7% (\u003cstrong\u003eFig. 4a, b, Supplementary Fig. 38\u003c/strong\u003e).\u0026nbsp;Oral gavage of LG/insulin-40 solution did not elicit obvious drops in BGLs, due to its swift dissociation and degradation in harsh gastric milieu (\u003cstrong\u003eSupplementary Fig. 39\u003c/strong\u003e).\u0026nbsp;Correspondingly, for mice receiving s.c. insulin, the plasma insulin concentration rapidly increased within 1 h, but then decreased in the following 2 h (\u003cstrong\u003eFig. 4c\u003c/strong\u003e). In contrast, the plasma exposure of insulin produced by LG/insulin-40-loaded capsules peaked at 4 h and was detectable over 8 h postdosing, yielding a relative oral bioavailability of 28.3% (\u003cstrong\u003eFig. 4d\u003c/strong\u003e). For groups treated with LG/insulin-40 solution, or blank or free insulin-loaded capsules,\u0026nbsp;BGLs and plasma insulin levels remained basal throughout.\u003c/p\u003e\n\u003cp\u003eWe then tracked the passage of LG/insulin-40, where insulin was pre-tagged with Cy5.5 (insulin\u003csup\u003eCy5.5\u003c/sup\u003e), in the intestinal tract. After oral gavage of LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40-loaded capsules, fluorescence signals propagated from the duodenum to jejunum, ileum, and colon over time, while the intensity of each segment firstly increased and then gradually weakened (\u003cstrong\u003eFig. 4e\u003c/strong\u003e). Cryosections of each lighted segment revealed prominent fluorescence signals inside the villi, evidencing that LG/insulin-40 crossed the mucosa irrespective of the intestinal regions. The permeation efficiency of LG/insulin-40 across different intestinal regions were evaluated \u003cem\u003eex vivo\u003c/em\u003e using an Ussing chamber system, which yielded \u003cem\u003eP\u003c/em\u003e\u003csub\u003eapp\u003c/sub\u003e values in the same order of magnitude (10\u003csup\u003e–4\u003c/sup\u003e) without significant differences\u0026nbsp;(\u003cstrong\u003eSupplementary Fig. 40\u003c/strong\u003e). Addition of DTNB or DTNB-loaded fusogenic liposomes\u0026nbsp;in the donor chamber significantly curtailed the absorptive transport of LG/insulin-40 (\u003cstrong\u003eSupplementary Fig. 41\u003c/strong\u003e).\u0026nbsp;These results demonstrated that the intestinal uptake of LG/insulin-40 hinged on the DCDE initiated by the thiols in the mucosa, while no obvious preferential absorption was biased towards one segment.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe also studied the distribution of signals of insulin\u003csup\u003eCy5.5\u003c/sup\u003e in major organs (\u003cstrong\u003eFig. 4f\u003c/strong\u003e),\u0026nbsp;which revealed that fluorescence signals appeared in the liver 1 h after oral dosing and reached the maximum after 4 h. At each post-gavage time point, the\u0026nbsp;fluorescence intensity was relatively stronger in the liver, followed by kidneys, but remained low in the heart, spleen, and lungs. This biodistribution pattern was concordant with the dynamics of endogenous insulin which is firstly transported to the liver for the ‘first pass’ and cleared by kidneys at the end\u003csup\u003e39\u003c/sup\u003e. Notably, fluorescence signals were still observed in the jejunum and ileum segments and the liver up to 8 h post administration. This was ascribed to that DCDE-mediated adhesion and accumulation of LG/insulin-40 in the mucus prolonged its retention in the gut, which thereby led to continued absorption of insulin and much longer hyperglycemic effect than s.c. insulin (\u003cstrong\u003eFig. 4a, c\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eWe next assessed the integrity of LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40 during the intestinal absorption and liver delivery. From the cryosection imaging, most of the LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40 remained as puncta in the epithelia, while few LG/insulin\u003csup\u003eCy5.5\u003c/sup\u003e-40scattered in the perisinusoidal space, and strongly diffusive fluorescence was recorded spreading throughout the hepatocytes (\u003cstrong\u003eFig. 4g, Supplementary Fig. 43\u003c/strong\u003e), indicating that LG/insulin-40 underwent dissociation mainly in the liver and less during enteral absorption.\u0026nbsp;As sinusoidal glutathione efflux is accompanied by a high local concentration of glutathione (7-10 mM)\u003csup\u003e40\u003c/sup\u003e,\u0026nbsp;we reasoned\u0026nbsp;that when the absorbed LG/insulin-40 was transited through the portal vein to hepatic sinusoids and diffused through the fenestrations to the perisinusoidal space, the poly(disulfide)s underwent\u0026nbsp;glutathione-triggered reductive depolymerization to liberate insulin in the liver. This hypothesis was cross-validated by the much faster\u0026nbsp;LG/insulin-40 dissociation kinetics at the hepatic glutathione pool level than those at villus epithelial and plasma glutathione levels (\u003cstrong\u003eSupplementary Fig. 32\u003c/strong\u003e).\u0026nbsp;Moreover, measurement of hepatic glycogen contents indicated that, compared to untreated and s.c. insulin-treated groups, hepatic glucose uptake in diabetic mice was significantly improved by LG/insulin-40 and became comparable to that of healthy mice over 7-day consecutive treatment (\u003cstrong\u003eFig. 4h\u003c/strong\u003e). Therefore, this hepatic glutathione pool-triggered insulin release would aid in restoring the liver as a primary metabolic regulator of glucose metabolism and correcting the aberrant glucose metabolism linked to s.c. insulin\u003csup\u003e41,42\u003c/sup\u003e, reinforcing the biopotency of LG/insulin-40 in mimicking endogenous insulin.\u003c/p\u003e\n\u003cp\u003eWe further evaluated the long-term biosafety of this oral insulin formulation \u003cem\u003evia\u003c/em\u003e dosing\u0026nbsp;diabetic\u0026nbsp;mice with LG/insulin-40-loaded capsules twice daily for two weeks. Like control diabetic mice receiving saline, all experimental mice\u0026nbsp;showed no dramatic fluctuations in body weight (\u003cstrong\u003eSupplementary Fig. 43\u003c/strong\u003e). Histological examination\u0026nbsp;of intestinal segments revealed unaltered mucosal thickness, intact\u0026nbsp;fingerlike villi, insignificant infiltration of immune cells and no opening of tight junctions (\u003cstrong\u003eSupplementary Figs. 44, 45\u003c/strong\u003e). The serum levels of endotoxin and typical proinflammatory cytokines and blood biochemical parameters in the experimental group were not statistically different from those of the control group (\u003cstrong\u003eSupplementary Figs. 4\u003c/strong\u003e\u003cstrong\u003e6-48\u003c/strong\u003e). The histological sections of each major organ displayed similar appearance in both groups with no discernible differences in pathological changes (\u003cstrong\u003eSupplementary Fig. 49\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGlycemic control in large animals\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe moved on to test the feasibility of LG/insulin-40 for perorally delivering insulin in large animals, aiming to bridge the gap between rodent models and human diabetes. As pigs\u0026nbsp;have similarities with\u0026nbsp;humans in\u0026nbsp;dietary needs, gastrointestinal physiology, and metabolism\u003csup\u003e43\u003c/sup\u003e, we leveraged streptozotocin-induced diabetic Bama minipigs as the large animal model (\u003cstrong\u003eFig. 5a, Supplementary Fig. 50\u003c/strong\u003e). Unlike s.c. insulin (0.5 IU kg\u003csup\u003e-1\u003c/sup\u003e) which produced a rapid hypoglycemic onset and quick resolution, oral gavage of LG/sinulin-40-loaded enteric capsules (size 2, \u003cstrong\u003eSupplementary Fig. 37\u003c/strong\u003e; 2 IU kg\u003csup\u003e-1\u003c/sup\u003e) led to a gradual decline in BGLs and longer hypoglycemic effect (\u003cstrong\u003eFig. 5b, Supplementary Fig. 51\u003c/strong\u003e), and yielded a pharmacological availability of 19.3% (\u003cstrong\u003eFig. 5c\u003c/strong\u003e). Meanwhile, a slower plasma insulin clearance rate was detected for capsule-delivered LG/insulin-40 compared to s.c. insulin (\u003cstrong\u003eFig. 5d\u003c/strong\u003e), producing a relative bioavailability of 20.8% (\u003cstrong\u003eFig. 5e\u003c/strong\u003e). Oral gavage of LG/insulin-40 solution, free insulin-loaded capsules, or blank capsules did not attenuate BGLs or induce plasma exposure of insulin. The comparable pharmacodynamic activity and relative bioavailability achieved in diabetic minipigs and mice underscore the interspecies translatability of this poly(disulfide)s-formulated oral insulin.\u003c/p\u003e\n\u003cp\u003eWe further assessed the efficacy of multi-day oral dosing in controlling postprandial and daily glycemia, as would be necessary for mitigating risks of diabetic cardiovascular complications\u003csup\u003e44\u003c/sup\u003e. Diabetic minipigs were treated with LG/insulin-40-loaded capsules (4 IU kg\u003csup\u003e-1\u003c/sup\u003e) 2 h before the two daily meals for two weeks (\u003cstrong\u003eFig. 5f, Supplementary Fig. 52\u003c/strong\u003e). Although control minipigs receiving preprandial s.c. insulin (1 IU kg\u003csup\u003e-1\u003c/sup\u003e, injected 0.5 h before each meal) showed severely suppressed postprandial BGLs, hypoglycemic states (\u0026lt;50 mg\u0026nbsp;dl\u003csup\u003e-1\u003c/sup\u003e) often occurred and BGLs restored hyperglycemic (\u0026gt;200 mg\u0026nbsp;dl\u003csup\u003e-1\u003c/sup\u003e) around 3-5 h after injection. By contrast, in at least two-thirds of cases, the oral intake of LG/insulin-40 controlled the postprandial glycemia below 200 mg\u0026nbsp;dl\u003csup\u003e-1\u003c/sup\u003e and further sustain the BGL within the range of 50-100 mg\u0026nbsp;dl\u003csup\u003e-1\u003c/sup\u003e for 1-4 h after the postprandial rise. Overall, the oral insulin maintained the BGL below 200 mg\u0026nbsp;dl\u003csup\u003e-1\u003c/sup\u003e for approximately 70% of the experimental period, with only a few short-lived instances of hypoglycemia. Analysis of the mean amplitude of glucose excursions (MAGEs) revealed much smaller daily fluctuations of BGLs in the LG/insulin-40 group than those in the s.c. insulin group, evidencing that minipigs intaking oral insulin experienced less glycemic variability (\u003cstrong\u003eFig. 5g, Supplementary Fig. 53\u003c/strong\u003e). This was further supported by the 1.9-fold reduction in serum glycated albumin levels, an indicator reflective of the glycemic control over the preceding 2-3 weeks\u003csup\u003e45\u003c/sup\u003e, in the oral insulin group (from 32.0% to 23.1%) than that of the s.c. insulin group (from 31.6% to 27.1%; \u003cstrong\u003eFig. 5h\u003c/strong\u003e). The better\u0026nbsp;management\u0026nbsp;of glycemia attained by oral insulin could be attributed to the liver-targeted delivery paradigm, which closely recreates the physiologic response to insulin under feeding conditions and provides better control over glucose metabolism across the liver and whole body\u003csup\u003e41,42\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThroughout the two-week consecutive treatment, minipigs treated with oral insulin behaved normally and showed neither alternation in feeding or stooling patterns nor reduction in body weight (\u003cstrong\u003eSupplementary Fig. 54\u003c/strong\u003e). At the end, histological analysis of different intestinal segments revealed normal morphology and no signs of inflammation (\u003cstrong\u003eSupplementary Fig. 55\u003c/strong\u003e). No obvious abnormalities in the blood biochemistry, complete blood count, and histology of major organs between the oral insulin and subcutaneous insulin treated groups were detected (\u003cstrong\u003eSupplementary Figs. 56-58\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eScale-up preparation and formulation generality\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the simplicity of this oral insulin formulation strategy, we further explored the feasibility of scaling up the formulation volume from 1 ml to 1 liter. LG/insulin-40 prepared in the larger batch presented similar physicochemical characteristics and attained a similar pharmacological availability in diabetic mice (\u003cstrong\u003eSupplementary Fig. 59\u003c/strong\u003e).\u0026nbsp;Of note, the disulfide monomer can be synthesized straightforwardly using inexpensive commercial chemicals and laboratory accessible instruments. As estimated, synthesis of 1 gram of LGs costs approximately 67.4 US$, which can be used to formulate about 9,100 IU of oral insulin; if a diabetic patient took this oral insulin at a dosage of 100 IU per time, the needed excipient monomers would cost about 0.7 US$\u0026nbsp;(\u003cstrong\u003eSupplementary Table S1\u003c/strong\u003e). Therefore, this oral insulin would render effective diabetes management at an economically reasonable level.\u003c/p\u003e\n\u003cp\u003eTo demonstrate the\u0026nbsp;adaptability\u0026nbsp;of this formulation strategy, we first chose HSA (isoelectric point ≈ 5, M.W. ≈ 67 kDa), a protein widely studied as a vector for various therapeutics,\u003csup\u003e46\u003c/sup\u003e as a model biomacromolecule drug. After dosing HSA/poly(disulfide)s complex\u0026nbsp;coacervates to 4T1 tumor-bearing mice, gradual accumulation of HSA in the tumor was observed over time (\u003cstrong\u003eSupplementary Figs. 60-62\u003c/strong\u003e). To further showcase the flexibility in choosing lipoic acid derivatives, we then used alendronate-conjugated lipoic acid as the monomer to coacervate sCT (isoelectric point = 8.86), a model peptide showing positive charges at physiological pH. When sCT/poly(disulfide)s complex\u0026nbsp;coacervates were ingested by mice, the hypocalcemia effect lasted longer compared to s.c. sCT, yielding a relative bioactivity of 22.8% (\u003cstrong\u003eSupplementary Figs. 63-69\u003c/strong\u003e).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe have developed\u0026nbsp;a simple yet effective strategy for\u0026nbsp;formulating oral biologics \u003cem\u003evia\u003c/em\u003e utilizing \u003cem\u003ein-situ\u003c/em\u003e polymerized poly(disulfide)s as the sole excipient. The systemic uptake of these biologics is mediated by the DCDE initiated by intrinsic thiols at different \u0026lsquo;stations\u0026rsquo; along the bodily journey of poly(disulfide)s-coacervated\u0026nbsp;biomacromolecules\u0026nbsp;(summarized in \u003cstrong\u003eFig. 6\u003c/strong\u003e). Using insulin as a model biologic drug and lipoic acid-guanidine derivative as the poly(disulfide)s monomer, we systematically studied how the complex coacervates were formulated and overcame the intestinal barriers without breaching them. While DCDE-mediated oral delivery of poly(disulfide)s-coacervated biomacromolecules is introduced for the first time, the remarkable pharmacodynamic and pharmacokinetic outcomes attained in murine and swine models, along with the successful scale-up preparation and the generality of this formulation approach, support the reliability of poly(disulfide)s-engineered oral biologics. As the rapid, dynamic nature of DCDE makes itself hard to characterize, these findings justify the need for gathering more detailed mechanical insights \u003cem\u003evia\u003c/em\u003e advanced technologies with higher resolutions in future studies.\u003c/p\u003e\n\u003cp\u003eImportantly, the dosing paradigm of poly(disulfide)s-formulated oral biologics offers the following advantages. Administration with patient-familiar capsules would afford greater convenience and medication compliance, obviating bowel discomfort or phobia of latent gastrointestinal perforation or obstruction. The ubiquitous existence of thiols in mucins and epithelial membranes could enable biomacromolecule absorption at different intestinal segments, providing prolonged systemic uptake and adequate bioavailability while avoiding limited or delayed uptake associated with specific receptor or region-targeting delivery systems. The noninvasive absorption procedure would leave the intestinal barriers intact, averting potential complications accompanying the manipulation of digestive enzyme activities, the mucus thickness, and the epithelial integrity \u003cem\u003evia\u003c/em\u003e chemical or physical approaches.\u0026nbsp;Still, local and systemic chronic effects over longer repeated dosing regimens require further study in subsequent preclinical and clinical assessments.\u003c/p\u003e\n\u003cp\u003eBy design, it can be naturally envisioned that this oral formulation is modular and scalable. In addition to the recombinant human insulin, insulin or insulin analogs with different therapeutic windows and half-lives should also be compatible with this poly(disulfide)s formulating strategy. Additionally, this formulation scheme could be expanded to develop a gallery of oral biologic drugs, which are currently administered by parenteral routes. From a chemical viewpoint, this is made possible by choosing appropriate lipoic acid derivatives depending on the interactions harnessed to interface the biomacromolecules. Overall, the simplicity, high bioavailability, scalability, and cost-effectiveness of poly(disulfide)s-formulated biologics would inspire the development of a new generation of oral therapeutics.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was funded by the National Key Research and Development Program of China (2020YFA0210800, H.H.Y.; 2022YFE0202200, Z.G.), the Major Project of Science and Technology of Fujian Province (2020HZ06006, H.H.Y.), the National Natural Science Foundation of China (22027805, H.H.Y.; 22334004, H.H.Y.; 22107019, Z.W.C.; 22277011, Z.W.C.), Start-up packages of Zhejiang University (Z.G.), and the Kunpeng Program grant (Z.G.).\u0026nbsp;Work performed at the Center for Nanoscale Materials, a U.S. Department of Energy Office of Science User Facility, was supported by the U.S. DOE, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eC.C., Z.W.C. and H.H.Y. designed the project. C.C., T.J.H., Z.L., H.J.L., Y.Z., X.L., Y.H.G., J.Y.L. and Z.W.C. performed the experiments. C.C., T.J.H., Z.L., H.J.L., Y.Z., X.L., Y.H.G., J.Y.L., T.W., W.H., H.C., E.A.R., G.J.C., Z.T.C., Z.W.C., Z.G. and H.H.Y. analyzed the data. C.C. and Z.W.C. drafted the original manuscript. C.C., Z.L., H.J.L., E.A.R., G.J.C., Z.W.C., Z.G. and H.H.Y. revised the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eC.C., T.J.H., Z.L, Z.W.C. and H.H.Y. are co-inventors on multiple patent applications covering aspects of the technology presented here. Z.G. is a scientific cofounder of ZCapsule Inc., Zenomics Inc., and μZen Pharma Inc. The other authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data generated or analyzed during this study are included in this published article (and its Supplementary Information files) or are available from the authors upon request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAbramson, A. et al. An ingestible self-orienting system for oral delivery of macromolecules. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e363\u003c/strong\u003e, 611-615 (2019).\u003c/li\u003e\n \u003cli\u003eHarrison, G.A. Insulin in alcoholic solution by the mouth. \u003cem\u003eBr. Med. J.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1204 (1923).\u003c/li\u003e\n \u003cli\u003eSims, E.K., Carr, A.L.J., Oram, R.A., DiMeglio, L.A. \u0026amp; Evans-Molina, C. 100 years of insulin: celebrating the past, present and future of diabetes therapy. \u003cem\u003eNat. Med.\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 1154-1164 (2021).\u003c/li\u003e\n \u003cli\u003eAbramson, A. et al. Oral delivery of systemic monoclonal antibodies, peptides and small molecules using gastric auto-injectors. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 103-109 (2022).\u003c/li\u003e\n \u003cli\u003eChen, Z. et al. Synthetic beta cells for fusion-mediated dynamic insulin secretion. \u003cem\u003eNat. Chem. Biol.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 86-93 (2018).\u003c/li\u003e\n \u003cli\u003eDrucker, D.J. Advances in oral peptide therapeutics. \u003cem\u003eNat. Rev. Drug. Discov.\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 277-289 (2020).\u003c/li\u003e\n \u003cli\u003eIyer, H., Khedkar, A. \u0026amp; Verma, M. Oral insulin \u0026ndash; a review of current status. \u003cem\u003eDiabetes Obes. Metab.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 179-185 (2010).\u003c/li\u003e\n \u003cli\u003eBrown, T.D., Whitehead, K.A. \u0026amp; Mitragotri, S. Materials for oral delivery of proteins and peptides. \u003cem\u003eNat. Rev. Mater.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 127-148 (2020).\u003c/li\u003e\n \u003cli\u003eLee, J.S. et al. Metabolic and immunomodulatory control of type 1 diabetes via orally delivered bile-acid-polymer nanocarriers of insulin or rapamycin. \u003cem\u003eNat. Biomed. Eng.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 983-997 (2021).\u003c/li\u003e\n \u003cli\u003eXiao, Y. et al. Oral Insulin Delivery Platforms: Strategies To Address the Biological Barriers. \u003cem\u003eAngew. Chem. Int. Ed.\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 19787-19795 (2020).\u003c/li\u003e\n \u003cli\u003eAnselmo, A.C., Gokarn, Y. \u0026amp; Mitragotri, S. Non-invasive delivery strategies for biologics. \u003cem\u003eNat. Rev. Drug. Discov.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 19-40 (2019).\u003c/li\u003e\n \u003cli\u003eHan, X. et al. Zwitterionic micelles efficiently deliver oral insulin without opening tight junctions. \u003cem\u003eNat. Nanotechnol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 605\u0026ndash;614 (2020).\u003c/li\u003e\n \u003cli\u003eLamson, N.G., Berger, A., Fein, K.C. \u0026amp; Whitehead, K.A. Anionic nanoparticles enable the oral delivery of proteins by enhancing intestinal permeability. \u003cem\u003eNat. Biomed. Eng.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 84-96 (2020).\u003c/li\u003e\n \u003cli\u003eBanerjee, A. et al. Ionic liquids for oral insulin delivery. \u003cem\u003eProc. Natl. Acad. Sci. USA\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, 7296 (2018).\u003c/li\u003e\n \u003cli\u003ePridgen Eric, M. et al. Transepithelial Transport of Fc-Targeted Nanoparticles by the Neonatal Fc Receptor for Oral Delivery. \u003cem\u003eSci. Transl. Med.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 213ra167-213ra167 (2013).\u003c/li\u003e\n \u003cli\u003eXiao, Y. et al. Glucose-responsive oral insulin delivery platform for one treatment a day in diabetes. \u003cem\u003eMatter\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 3269-3285 (2021).\u003c/li\u003e\n \u003cli\u003eHub\u0026aacute;lek, F. et al. Molecular engineering of safe and efficacious oral basal insulin. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 3746 (2020).\u003c/li\u003e\n \u003cli\u003eAbramson, A. et al. A luminal unfolding microneedle injector for oral delivery of macromolecules. \u003cem\u003eNat. Med.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 1512-1518 (2019).\u003c/li\u003e\n \u003cli\u003eBanerjee, A., Lee, J. \u0026amp; Mitragotri, S. Intestinal mucoadhesive devices for oral delivery of insulin. \u003cem\u003eBioeng. Transl. Med.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 338-346 (2016).\u003c/li\u003e\n \u003cli\u003eBang, E.-K. et al. Substrate-Initiated Synthesis of Cell-Penetrating Poly(disulfide)s. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e135\u003c/strong\u003e, 2088-2091 (2013).\u003c/li\u003e\n \u003cli\u003eZhou, J. et al. Self-Assembled and Size-Controllable Oligonucleotide Nanospheres for Effective Antisense Gene Delivery through an Endocytosis-Independent Pathway. \u003cem\u003eAngew. Chem. Int. Ed.\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e, 5236-5240 (2019).\u003c/li\u003e\n \u003cli\u003eGasparini, G. et al. Cellular Uptake of Substrate-Initiated Cell-Penetrating Poly(disulfide)s. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e136\u003c/strong\u003e, 6069-6074 (2014).\u003c/li\u003e\n \u003cli\u003eQian, L. et al. Intracellular Delivery of Native Proteins Facilitated by Cell-Penetrating Poly(disulfide)s. \u003cem\u003eAngew. Chem. Int. Ed.\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 1532-1536 (2018).\u003c/li\u003e\n \u003cli\u003eLaurent, Q. et al. Thiol-Mediated Uptake. \u003cem\u003eJACS Au\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 710-728 (2021).\u003c/li\u003e\n \u003cli\u003eTorres, A.G. \u0026amp; Gait, M.J. Exploiting cell surface thiols to enhance cellular uptake. \u003cem\u003eTrends Biotechnol.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 185-190 (2012).\u003c/li\u003e\n \u003cli\u003eCheng, Y. et al. Inhibitors of thiol-mediated uptake. \u003cem\u003eChem. Sci.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 626-631 (2021).\u003c/li\u003e\n \u003cli\u003eArtemova, N., Stein-Margolina, V., Smirnova, E. \u0026amp; Gurvits, B. Formation of supramolecular structures of a native-like protein in the presence of amphiphilic peptides: Variations in aggregate morphology. \u003cem\u003eFEBS Lett.\u003c/em\u003e \u003cstrong\u003e586\u003c/strong\u003e, 186-190 (2012).\u003c/li\u003e\n \u003cli\u003eHubatsch, I., Ragnarsson, E.G.E. \u0026amp; Artursson, P. Determination of drug permeability and prediction of drug absorption in Caco-2 monolayers. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 2111-2119 (2007).\u003c/li\u003e\n \u003cli\u003eHalberg, I.B. et al. Efficacy and safety of oral basal insulin versus subcutaneous insulin glargine in type 2 diabetes: a randomised, double-blind, phase 2 trial. \u003cem\u003eLancet Diabetes Endocrinol.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 179-188 (2019).\u003c/li\u003e\n \u003cli\u003eDi Pisa, M., Chassaing, G. \u0026amp; Swiecicki, J.-M. Translocation Mechanism(s) of Cell-Penetrating Peptides: Biophysical Studies Using Artificial Membrane Bilayers. \u003cem\u003eBiochemistry\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 194-207 (2015).\u003c/li\u003e\n \u003cli\u003eGasparini, G., Bang, E.-K., Montenegro, J. \u0026amp; Matile, S. Cellular uptake: lessons from supramolecular organic chemistry. \u003cem\u003eChem. Commun.\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 10389-10402 (2015).\u003c/li\u003e\n \u003cli\u003eDerivery, E., Bartolami, E., Matile, S. \u0026amp; Gonzalez-Gaitan, M. Efficient Delivery of Quantum Dots into the Cytosol of Cells Using Cell-Penetrating Poly(disulfide)s. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e139\u003c/strong\u003e, 10172-10175 (2017).\u003c/li\u003e\n \u003cli\u003eGuo, M. et al. Probing the Stochastic, Motor-Driven Properties of the Cytoplasm Using Force Spectrum Microscopy. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e158\u003c/strong\u003e, 822-832 (2014).\u003c/li\u003e\n \u003cli\u003eNkabyo, Y.S., Ziegler, T.R., Gu, L.H., Watson, W.H. \u0026amp; Jones, D.P. Glutathione and thioredoxin redox during differentiation in human colon epithelial (Caco-2) cells. \u003cem\u003eAm. J. Physiol. - Gastrointest. Liver Physiol.\u003c/em\u003e \u003cstrong\u003e283\u003c/strong\u003e, G1352-G1359 (2002).\u003c/li\u003e\n \u003cli\u003ePulcu, G.S. et al. Single-Molecule Kinetics of Growth and Degradation of Cell-Penetrating Poly(disulfide)s. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e141\u003c/strong\u003e, 12444-12447 (2019).\u003c/li\u003e\n \u003cli\u003eChuard, N., Gasparini, G., Roux, A., Sakai, N. \u0026amp; Matile, S. Cell-penetrating poly(disulfide)s: the dependence of activity, depolymerization kinetics and intracellular localization on their length. \u003cem\u003eOrg. Biomol. Chem\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 64-67 (2015).\u003c/li\u003e\n \u003cli\u003eGum, J.R. et al. The human MUC2 intestinal mucin has cysteine-rich subdomains located both upstream and downstream of its central repetitive region. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cstrong\u003e267\u003c/strong\u003e, 21375-21383 (1992).\u003c/li\u003e\n \u003cli\u003eBansil, R. \u0026amp; Turner, B.S. The biology of mucus: Composition, synthesis and organization. \u003cem\u003eAdv. Drug Deliv. Rev.\u003c/em\u003e \u003cstrong\u003e124\u003c/strong\u003e, 3-15 (2018).\u003c/li\u003e\n \u003cli\u003eTokarz, V.L., MacDonald, P.E. \u0026amp; Klip, A. The cell biology of systemic insulin function. \u003cem\u003eJ. Cell Biol.\u003c/em\u003e \u003cstrong\u003e217\u003c/strong\u003e, 2273-2289 (2018).\u003c/li\u003e\n \u003cli\u003eJiang, X., Du, B. \u0026amp; Zheng, J. Glutathione-mediated biotransformation in the liver modulates nanoparticle transport. \u003cem\u003eNat. Nanotechnol.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 874-882 (2019).\u003c/li\u003e\n \u003cli\u003eEdgerton, D.S. et al. Targeting insulin to the liver corrects defects in glucose metabolism caused by peripheral insulin delivery. \u003cem\u003eJCI Insight\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, e126974 (2019).\u003c/li\u003e\n \u003cli\u003eLewis, G.F., Carpentier, A.C., Pereira, S., Hahn, M. \u0026amp; Giacca, A. Direct and indirect control of hepatic glucose production by insulin. \u003cem\u003eCell Metab.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 709-720 (2021).\u003c/li\u003e\n \u003cli\u003eKleinert, M. et al. Animal models of obesity and diabetes mellitus. \u003cem\u003eNat. Rev. Endocrinol.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 140-162 (2018).\u003c/li\u003e\n \u003cli\u003eBrown, A., Reynolds, L.R. \u0026amp; Bruemmer, D. Intensive glycemic control and cardiovascular disease: an update. \u003cem\u003eNat. Rev. Cardiol.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 369-375 (2010).\u003c/li\u003e\n \u003cli\u003eFang, M., Daya, N., Coresh, J., Christenson, R.H. \u0026amp; Selvin, E. Glycated Albumin for the Diagnosis of Diabetes in US Adults. \u003cem\u003eClin. Chem.\u003c/em\u003e \u003cstrong\u003e68\u003c/strong\u003e, 413-421 (2022).\u003c/li\u003e\n \u003cli\u003eHoogenboezem, E.N. \u0026amp; Duvall, C.L. Harnessing albumin as a carrier for cancer therapies. \u003cem\u003eAdv. Drug Deliv. Rev.\u003c/em\u003e \u003cstrong\u003e130\u003c/strong\u003e, 73-89 (2018).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"drug delivery, oral biologics, dynamic covalent chemistry, poly(disulfide)s, complex coacervates, diabetes treatment","lastPublishedDoi":"10.21203/rs.3.rs-3616020/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3616020/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe biological barriers present in the intestine thwart the absorption of orally delivered biologics, which, if overcome, would reduce the injection burdens for millions of patients. Here, we present a straightforward yet effective oral biologic formulation, which utilizes \u003cem\u003ein-situ\u003c/em\u003e growing poly(disulfide)s as the sole excipient to circumvent all intestinal barriers in a noninvasive way. We find that, through dynamic covalent disulfide exchange initiated by the thiols in mucins, epithelial membranes, and hepatic sinusoids, digestion-resistant complex coacervates formed from insulin (as a model drug) and guanidinium-containing poly(disulfide)s readily traverse the mucus and epithelial layers without remodeling the barriers’ integrity, and then undergo dissociation to release insulin in the liver. Oral gavage of the complex coacervates with enteric capsules into diabetic mice and swine models elicited a-few-hour longer hypoglycemic effect than subcutaneously injected insulin, attaining relative bioavailabilities over 20%. 14-day dosing experiments demonstrated the postprandial and daily glycemic control capacity and the biosafety of this oral insulin. The generality of this formulation scheme was further validated with human serum albumin and salmon calcitonin.\u003c/p\u003e","manuscriptTitle":"Dynamic Covalent Disulfide Exchange Mediates Oral Delivery of Biomacromolecules","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-08 18:48:17","doi":"10.21203/rs.3.rs-3616020/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"abfeadd7-8ec5-4f4c-ab54-0140441ab821","owner":[],"postedDate":"March 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":27931324,"name":"Biological sciences/Drug discovery/Drug delivery"},{"id":27931325,"name":"Health sciences/Medical research/Drug development"}],"tags":[],"updatedAt":"2025-06-16T12:31:20+00:00","versionOfRecord":[],"versionCreatedAt":"2024-03-08 18:48:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3616020","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3616020","identity":"rs-3616020","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}