123 - Feasibility Studies in the Nascent Field of in Vivo Proteomics: Implications for a New Domain for IR?

Purpose: Protein expression is dynamic and context-dependent, yet it is mainly studied in vitro. We sought to determine feasibility to detect putative differences in protein expression in vitro and in vivo and evaluate if IR techniques can be leveraged to determine and potentially manipulate protein expression profiles in vivo.

Materials and Methods: A 3-phase experiment was conducted. Phase 1 consisted of a 4-arm binary go/no go design with a negative control, tail vein injection (100 µL of iodoacetamide alkyne, IAA, Vector Labs, Plain City, OH), perfusion, and post-lysis positive control in 4 Dbh^cre KI male mice (Jackson Laboratory, Bar Harbor, ME). For perfusion IAA was delivered in vivo at 1 millimolar concentration in saline at 2.5 mL/min for 10 min via direct left ventricular puncture followed by perfusion with ice-cold PBS at pH 7.4. Tissues were harvested, homogenized, and lysed, and lysates treated with Cy5-azide (BroadPharm, San Diego CA) and CuSO₄ under click conditions to visualize labeling. SDS PAGE was performed on the reaction products and imaged using in-gel fluorescence. Phase 2 consisted of 3 arms with in vitro cells (Bnl Cl.2 liver cells), untreated liver (3 male Dbh^cre KI mice), and in vivo perfusion with 3 Dbh^cre KI mice (2M/1F). Lysates from cell culture and untreated liver were alkylated with IAA, then digested for proteomic analysis. LC-MS/MS was performed using a timsTOF Pro2 mass spectrometer (Bruker Daltonics, Billerica MA). Phase 3 consisted of data analysis. MS raw files were searched FragPipe GUI version 20 with the MSFragger (version 3.8) search algorithm. Protein identification used the UniProtKB/Swiss-Prot protein database for Mus musculus (release 2025_05; 17,052 entries). Only proteins found in at least 2 of the 3 subjects were included and data were purged of duplicate entries.

Results: Phase 1 fluorescence imaging showed a qualitatively strong positive signal for the perfusion and post-lysis positive control, and negligible signal for the negative control and the tail-vein injection. Phase 2 provided 3 distinct lysates and data sets for analysis. Phase 3 revealed 1,829 distinct proteins. Of 192 proteins in the perfusion arm, 34 or 18% were unique.

Conclusion: Detection of differences in protein expression in vitro and in vivo is clearly feasible and clear differences exist based on conditions. This implies that percutaneous delivery methods will be readily adapted and will likely prove to be powerful tools when brought to bear in discovery of new proteins and for manipulation in vivo across a wide range of conditions in the future.