Donor-Acceptor Pyridinium Salts for Photo-Induced Electron-Transfer-Driven Modification of Tryptophan in Peptides, Proteins, and Proteomes Using Visible Light

Abstract

Tryptophan (Trp) plays a variety of critical functional roles in protein biochemistry; however, owing to its low natural frequency and poor nucleophilicity, the design of effective methods for both single protein bioconjugation at Trp as well as for in situ chemoproteomic profiling remains a challenge. Here, we report a method for covalent Trp modification that is suitable for both scenarios by invoking photo-induced electron transfer (PET) as a means of driving efficient reactivity. We have engineered biaryl N-carbamoyl pyridinium salts that possess a donor-acceptor relationship that enables optical triggering with visible light whilst simultaneously attenuating the probe's photo-oxidation potential in order to prevent photodegradation. This probe was assayed against a small bank of eight peptides and proteins, where it was found that micromolar concentrations of the probe and short irradiation times (10-60 min) with violet light enabled efficient reactivity toward surface exposed Trp residues. The carbamate transferring group can be used to transfer useful functional groups to proteins including affinity tags and click handles. DFT calculations and other mechanistic analyses reveal correlations between excited state lifetimes, relative fluorescence quantum yields, and chemical reactivity. Biotinylated and azide-functionalized pyridinium salts were used for Trp profiling in HEK293T lysates and in situ in HEK293T cells using 440 nm LED irradiation. Peptide-level enrichment from live cell labeling experiments identified 290 Trp modifications, with 82% selectivity for Trp modification over other π-amino acids, demonstrating the ability of this method to identify and quantify reactive Trp residues from live cells.

Figure 1.

(A) Functional roles of the “tryptophan-ome” and chemical challenges of targeting tryptophan for chemoproteomic profiling. (B) Harnessing photo-induced electron transfer (PET) for Trp modification using N-carbamoyl pyridinium salts. (C) This work: Visible light activated probe for selective Trp modification in purified peptides and in situ proteomic profiling.

Figure 2.

(A) Design, absorption/emission, fluorescence lifetimes, and calculated redox potentials of Trp probe 2a. Calculations were performed using B3LYP 6–31g(d) using an H₂O solvent cavity. (B) Labelling of lysozyme with 2a. Cyan residues=Trp, Red residues=Tyr. (C) Stern-Volmer plot of fluorescence quenching of 2a with NATA. (D) Establishment of temporal control with a lamp “on-off” experiment using identical conditions to those shown in Fig. 2B. (E) Mechanistic considerations for the requirement of proximity of [2a]* to the protein of interest. ^aConversions are estimated by total ion count (TIC) and represent the average of two experiments.

Figure 3.

(A) Exploration of the labelling of Trp residues in various purified peptides and proteins using 2a. Cyan residues=Trp, Red residues=Tyr. (B) Calculated solvent accessibilities for all modified and unmodified Trp residues except that of Octreotide are highlighted in the boxplot. (C) Transferring group scope using the 2 scaffold. ^aConversions estimated by (TIC) and represent the average of two experiments. ^bModification sites confirmed by MS-MS.

Figure 4.

(A) Evaluation of three Trp probe designs against HEK293T lysates. All experiments were performed in duplicate. (B) Post-enrichment elution profiles of 1a, 1b, and 2b at 0–100 μM probe and at 100 μM probe in the absence of light. (C) Number of proteins identified at 100 μM of each probe. (D) Venn diagram comparing overlap of proteome coverage by each probe. (E) Classes of proteins showing significant enrichment relative to all detected proteins with 100 μM 2b. −Log₁₀ p-values are shown to clearly highlight enrichment. (F) Light dependence of protein-level enrichment with 2b.

Figure 5.

(A) Workflow for peptide level enrichment of Trp in HEK293T cells using bifunctional probe 2d. (B) Volcano plots showing light-dependent and dose-responsive enrichment of the tryptophan-ome from 10–100 μM Trp. Average of two experiments. Light blue dots indicate enriched peptides harboring Trp modifications. Dark blue dots indicate labelled Trp in NPM1 and PARP1. (C) Detected residue modifications by percentage. (D) Chemoselectivity based upon amino acid relative frequency. (E) Subcellular localization of Trp-modified proteins.

Figure 6.

(A) Modification of functionally critical Trp residue in nuclear proteins NPM1 (Trp-290) and PARP1 (Trp-79 and Trp-318). Residues engaging in non-covalent interactions with both Trp residues PARP1 are highlighted. Functionally critical Trp-246, which has lower solvent accessibility and was not modified, is also highlighted. SA=solvent accessibility. (B) Comparison of solvent accessibility in modified and unmodified Trp residues of proteins identified in situ with 2d. (C) Modified Trp residues on proteins associated with various disease states.

Figure 7.

(A) Validation of the identification of NPM1 and PARP1 via western analysis of post-protein level enrichment profiles of HEK293T cells labelled with 2d. (B) Labelling of a recombinant NPM1 C-terminal construct with 2b.