UC Berkeley

Bioorthogonal chemistries are reactions that are designed to proceed in living environments without perturbing endogenous biological functionalities. These reactions are valuable tools for labeling and studying biomolecules both in vitro and in vivo, often providing unique insights into dynamic, living processes. For a reaction to be considered bioorthogonal, it must proceed in aqueous solvents at physiological pH and temperature. The reaction must also be rapid and selective, generating a stable, covalent adduct that is not reactive towards biological functionalities. Finally, one of the reaction partners must be capable of installation onto the biomolecule of interest.

A major motivator in the development of bioorthogonal chemistries is their potential utility in imaging and studying biomolecules in living animals. Chapter one chronicles advancements in the use of bioorthogonal reactions to tag biomolecules in multicellular organisms, focusing on the most prevalent reactions developed to date -- the Staudinger ligation, copper-click chemistry, copper-free click chemistry, and the tetrazine ligation. Examples are provided to highlight the importance of fast reaction kinetics as well as pharmacokinetics on the success of a ligation in vivo. Chapter one also provides commentary on unmet challenges in the field as well as an outlook on future advancements.

The in vivo applications of bioorthogonal chemistry discussed in chapter one serve as motivation for the experimental work presented in chapter two. Here, we describe our efforts to understand the factors that contribute to the kinetic profile of the copper-free click reaction. Copper-free click chemistry is a bioorthogonal 1,3-dipolar cycloaddition between azides and strained cyclooctynes to form triazoles. The reaction has seen widespread use in selectively tagging biomolecules both in vitro and in vivo. These successes have prompted the development of cyclooctyne analogs with improved reactivity toward the azide. However, predicting a cyclooctyne's reactivity is challenging, requiring researchers to design and undertake lengthy syntheses of alkynes that may or may not prove successful bioorthogonal reagents. In chapter two, we discuss our work towards defining and predicting the effects of strain and electronics on the reactivity of a cyclooctyne reagent. Through synthesis of analogs of biarylazacyclooctynone (BARAC), the fastest cyclooctyne developed to date, and subsequent reactivity measurements, we gain new insights into the effects of cyclooctyne strain and electronics on reactivity. As well, through computational modeling of our BARAC analogs we conclude that the distortion/interaction model of 1,3-dipolar cycloaddition kinetics serves as a valuable predictor of cyclooctyne reactivity in the copper-free click reaction.

Chapter three describes our motivation to develop new bioorthogonal ligations, highlighting the dearth of mutually orthogonal reactions capable of achieving multiplexed imaging. In addition, we discuss the need for bioorthogonal chemistries with new functional capabilities (i.e. polymerizations, reversible reactions, etc.). We then introduce the quadricyclane (QC) ligation, a new bioorthogonal reaction developed in the Bertozzi lab. The QC ligation is a formal [2s+2s+2p] reaction between QC and nickel bis(dithiolene). The reaction has been shown to fulfill many of the requirements of bioorthogonality, but no method of incorporating the QC functionality into a biomolecule of interest has been demonstrated. In chapter three, we discuss our use of the pyrrolysine synthetase/tRNACUA system for site-specific incorporation of a QC amino acid into a protein and subsequent tagging of this QC functionality with a nickel bis(dithiolene) reagent.

In chapter four we discuss efforts to further develop the QC ligation, exploring new chemical transformations accessible through this unique reaction. Specifically, we analyze the photodissociation of the QC/nickel bis(dithiolene) adduct to form nickel bis(dithiolene) and norbornadiene, a transformation that has the potential to make the QC ligation a "click-unclick" reaction. In addition, we have begun to analyze possible secondary reaction partners for the norbornadiene product of the photodissociation. Chapter four chronicles our ongoing work to optimize these unique chemical transformations for reversible tagging of model proteins.