Therapeutic agents including small molecules, peptides, and proteins comprise most of the pharmaceutics currently on the market and are an invaluable piece of modern medicine. Many of these therapeutics have some level of instability and/or off-target effects when administered in vivo that ultimately limits their effectiveness. Covalent modification of such therapeutics into an inactive prodrug prior to administration is a common modality used within the field of drug delivery to increase the circulation time or limit off-target effects, thus enhancing the therapeutic efficacy compared to administration of the unaltered drug. Due to the large diversity in chemical reactivities across various therapeutics, a wide array of strategies has been developed to covalently modify functional groups of interest. This thesis outlines the development of two such prodrug modalities; the first for the targeted delivery of oxycodone to the small intestine to prevent nonprescribed forms of administration and the second being a platform for the tunable release of peptides and proteins for enhanced circulation time.

Biologics including peptides, proteins, and oligonucleotides are an important class of macromolecules that have received an increasing amount of attention over the past few decades for their therapeutic potential. Their widespread adoption has lacked, in part, due to issues with chemical instability and immunogenicity in vivo. One particular avenue towards mitigating these effects is through PEGylation of the biologics, which can lead to enhanced stability and circulation times in vivo. However, PEGylation typically results in decreased binding or activity and in some cases can turn these off altogether. To mitigate these effects, researchers developed small molecule linkages that can be placed in between the protein and polymer, which cleave slowly over time to release native protein with restored binding/activity. Chapter 1 outlines the current strategies used within the field of traceless peptide/protein conjugation, including lonapegsomatropin-tcgp and NKTR-214 the only two traceless protein-polymer conjugates currently approved by the FDA, along with insights into potential future directions for this field. Prescription opioids, although necessary for pain management, are highly addictive and have led to the dramatic increase of opioid involved overdoses over the past 20 years. Abuse-deterrent (AD) opioid formulations are an important avenue towards addressing this national health crisis by increasing the difficulty for abusers to easily obtain large amounts of active opioid. However, there are currently only ten AD formulations on the market, a majority of which are easily circumvented by motivated users. Chapter 2 outlines the design and preparation of a slow-release elastomeric opioid formulation, which requires the presence of two proteases found within the small intestine in order to release the drug. This strategy limits the common abusive routes of administration, including, nasal insufflation and intravenous injection through the protease mediated activation step. In order to do this, the opioid was modified with a dual-enzyme responsive peptide sequence, whereupon proteolytic cleavage by chymotrypsin followed by trypsin triggers a self-immolative based release of the active opioid. The resulting elastomeric formulation was shown to be stable towards mechanical deformation testing following periods of heating and cooling, as well as hydrolytic degradation across a pH range of 2-10.

It was later identified that the elastomeric opioid carrier was not entirely necessary for the abuse deterrent properties of the formulation, but rather the same properties could be demonstrated in a small molecule peptide-opioid prodrug. Therein, Chapter 3 highlights the development of a second-generation abuse-deterrent peptide prodrug without the elastomeric carrier. The peptide-opioid prodrug sequence was optimized to mitigate any nonspecific protease cleavage, while also enhancing the rate of trypsin and chymotrypsin cleavage, the two enzymes responsible for activation in the small intestine. Notably, the composition of the amino acid sequence was highly crucial in determining the protease promiscuity which was identified as an issue with the first-generation prodrug. In addition, a third level of protection from abuse was incorporated into the prodrug through an acid mediated activation step to unmask a tyrosine residue. This activation rapidly occurs in simulated gastric fluid producing t-butanol and allows for chymotrypsin to bind and subsequently cleave the tyrosine residue.

The development of traceless linkers, outlined in Chapter 1, demonstrates the interest in creating prodrug like protein-polymer conjugates, allowing for the systemic administration of therapeutic peptides and proteins that were previously cleared too rapidly to be clinically useful. To this end, Chapter 4 outlines the development of a new amine reactive benzylamine traceless linker as an alternative to the commonly used carbamate linkages. A small molecule model system was initially used to probe the release kinetics of a primary amine, simulating lysine release from a protein-polymer conjugate. Varying the electronics within the aromatic core of the linker played a vital role in modulating the rate of release, resulting in half-lives ranging from 144 to 20 hours. The most promising linker was then incorporated into as a PEG-end group and used to conjugate to prepare traceless lysozyme-PEG conjugates. These conjugates released more than 95% native lysozyme over the course of 48 hours at a pH of 7.4 with a restoration of lysozyme activity upon release, whereas less than 25% native lysozyme was released within 96 hours at a pH of 4.0.

Building upon the results from Chapter 4 and mechanistic insights from density functional theory (DFT) calculations carried out in collaboration with members from the Houk Lab, Chapter 5 discusses the synthetic pursuits, kinetics studies, and protein conjugation studies toward developing a second-generation linker. These linkers incorporate an intramolecular trap to decrease the lifetime of the quinone methide intermediate, which in turn enhanced the rate of release to a 4.5 hour half-life using a small molecule model system. Additionally, these second-generation benzylamine linkers were used to prepare two traceless lysozyme-PEG conjugates with varying electronics and rates of release. The lysozyme-PEG conjugate containing the more electron rich linker demonstrated 98% release of native lysozyme within 12 days while restoring lysozyme activity, whereas the less electron rich linker showed only 50% release within the same time frame. This new class of linkers with tunable release rates expands the traceless linkers toolbox for a variety of bioconjugation applications.