PROJECT SUMMARY
Chemical modification of proteins is an important strategy for understanding their structure/function and
has guided the design of new pharmaceutical products and targeted therapeutics. Accordingly, new technologies
that target amino acid side chains in both selective and unselective fashions are necessary to gain advanced
biochemical information. In bioconjugation processes, amino acids are selectively targeted and modified.
However, classical bioconjugation strategies rely upon the inherent nucleophilicity of amino acid side chains,
rendering the selective modification of less nucleophilic sites a significant challenge. At the same time, the ability
to rapidly modify a variety of amino acids is necessary for proximity labeling, a strategy where proteins are tagged
with reactive probes for mapping the cellular microenvironment to better understand spatial connections and
relationships between biomolecules. Visible light photoredox catalysis has emerged as a powerful platform for
modifying amino acids through redox events. The use of visible light as a sustainable and non-invasive stimulus
not only facilitates the mild generation of reactive intermediates, but also enables spatiotemporal control over
chemical reactivity, a desirable feature particularly in the treatment of disease. Nevertheless, the extension of
photoredox catalysis into biological applications is underexplored because photocatalyst activation requires high-
energy blue light, which penetrates tissue poorly and can result in background activation of molecules.
This proposal describes the development and biological applications of red light bioconjugation and
proximity labeling platforms using Os(II) photocatalysts that proceed via the conversion of small molecules into
reactive probes that can react with amino acids selectively or promiscuously. A method for selective
bioconjugation of aspartic acid (Asp) and glutamic acid (Glu) is described wherein red light-photoexcited Os
photocatalysts activate vinyl azides into strained 2H-azirines, molecules that have been demonstrated to
undergo selective reactions with carboxylates instead of other more nucleophilic groups. A unique mechanism
is proposed upon reaction of a carboxylate group with photocatalytically-generated 2H-azirines to produce
oxazole-derived products in the covalent modification. A variety of Os catalysts with high triplet energy values
will be synthesized and evaluated in this method, which can then be translated to chemoselective labeling of
these residues on peptides and proteins. The ability of this class of photocatalysts to generate carbenes from
(trifluoromethyl)aryl diazos will also be explored. Reactivity will be established in the context of O–H insertion,
reactions with amino acids, and in isolated proteins. Due to their reactive nature, the carbenes react with a variety
of side chains rapidly, a feature that will be leveraged in tightening labeling radii in proximity labeling on cell
surfaces. Proximity labeling mediated by low energy red light presents two significant advances: (1) the use of
tissue-penetrating and noncytotoxic red light, and (2) minimization of any background substrate activation,
enabling high precision in spatiotemporal labeling.