Leveraging gut commensals to create a “clickable” drug delivery platform for oral corticosteroids

Introduction:: Oral corticosteroids are effective at treating mild-to-moderate colitis and other autoimmune disorders, but their long-term use is limited due to adverse effects upon prolonged systemic exposure. Existing strategies to increase the gut residence time of drugs largely rely on mucoadhesive biopolymers; however, these often afford only marginal improvements due to the rapid turnover rate of the intestinal mucosa. We have created a novel oral delivery paradigm that can improve both the specificity and retention of corticosteroids targeted to the gut. Our approach entails labeling the surface of gut commensals with a “clickable” azide group and subsequently delivering a corticosteroid prodrug containing the complementary strained cyclooctyne (here, DBCO). In this manner, prodrug that encounters labeled gut commensals (predominately found in the colon) spontaneously forms a covalent bond via the catalyst-free strain promoted azide-alkyne cycloaddition (SPAAC) reaction. Over time, this tethered prodrug degrades to locally release the unadulterated free corticosteroid whereas prodrug that traffics elsewhere (e.g., enters systemic circulation) is cleared before a substantial portion has had time to degrade. In this manner, our platform can improve both the specificity and increase the duration of action of corticosteroids, rendering them safe for longer term use.


Materials and Methods:: The budesonide prodrug bPD was synthesized by first combining NH2-PEG-COOH with an excess of DBCO-NHS in DCM containing DIEA. The crude product was washed with cold diethyl ether and dialyzed to remove unreacted DBCO-NHS (yield = 60 – 80%). The COOH-PEG-DBCO intermediate was then reacted with budesonide via a Steglich esterification in the presence of DCM, DMAP, and HoAt in anhydrous DCM under N2. The final product was purified via HPLC and lyophilized to yield a viscous liquid (yield = 20%). For the in vitro release assays, bPD was dissolved in either simulated gastric fluid (pH = 1.2), simulated colonic fluid (pH = 6.8), or simulated intestinal fluid (pH = 7.4) and periodically sampled via UPLC. The extent of degradation was determined by comparing the relative peak areas corresponding to the undegraded bPD and budesonide.

HeLa cells, which endogenously express human glucocorticoid receptor, were transfected with a plasmid that places control of EGFP under the MMTV promoter-enhancer and placed under selection for several days. The resulting polyclonal population was seeded at a density of 12,500 cells/well in a 96-well plate. The next day, the cell media was replaced with serum-free media containing 0, 0.01, 0.1, 1, or 5 μM undegraded bPD, fully degraded bPD, or free budesonide. After 1 hour, the cells were washed once with PBS and complete media was added to each well. 20 hours later, cells were washed, dissociated from the plate, resuspended in PBS, and assayed for EGFP expression using flow cytometry.


Results, Conclusions, and Discussions:: We synthesized a novel budesonide prodrug via the scheme shown in Fig. 1A. This scheme consists of an N-hydroxysuccinimide (NHS)/amine reaction to form the PEG-DBCO molecule, followed by a Steglich esterification to form a hydrolysable ester bond between the PEG-DBCO and budesonide. We showed that the resulting prodrug, budesonide-PEG-DBCO (bPD), is capable of fully degrading into free budesonide and PEG-DBCO (Fig. 1B), with a half-life ranging from 11 hours in simulated intestinal fluid to 25 hours in simulated gastric fluid (Fig. 1C) . To investigate the bioactivity of bPD, we engineered a HeLa cell line to stably contain a genetic construct that links activation of the mouse mammary tumor virus (MMTV) promoter-enhancer region, a known glucocorticoid response element, to expression of EGFP . Using this cell assay, we created a dose-response curve for bPD and free budesonide. As shown in Fig. 1D, undegraded bPD has an apparent EC50 that is 6-fold higher than that of the free drug (0.72 μM vs. 0.12 μM), whereas fully degraded bPD has an EC50 that is comparable to that of the free drug. This supports the notion that bPD is a prodrug of budesonide with reduced activity.

To determine the feasibility of using the SPAAC reaction to label the surface of gut commensals, we anaerobically cultured a fecal sample from a gut-humanized mouse ex vivo. Bacteria were co-incubated with either 1 mM azido-D-alanine (AzDA) or D-alanine (DA; negative control) for 24 hours and subsequently labeled with 5 μM Cy5-DBCO. As shown in Fig. 1E, AzDA was able to label effectively 100% of the bacteria as evidenced by a significant shift in fluorescence; in comparison, a minimal change in fluorescence was observed for the population co-incubated with DA.

These studies suggest that (1) bPD is a DBCO-containing prodrug capable of releasing free budesonide over physiologically-relevant timescales and (2) the SPAAC reaction can be used to label the surface of a diverse population of gut commensals with an accessible azide moiety. Taken together, our delivery strategy has the potential to improve the efficacy and safety profile of corticosteroids targeted to the gut.


Acknowledgements (Optional): : T.H.B. was supported by the National Science Foundation (grant no. 2236422).

References (Optional): :