Biased Agonism at CXCR3 Drives Differential Phosphoproteomic and Transcriptomic Profiles and Cellular Outputs

Objective To demonstrate that the endogenous agonists of the chemokine receptor CXCR3 differentially impact the phosphoproteome and transcriptome to drive distinct cellular outputs. Background G‐Protein Coupled Receptors (GPCRs) are the largest family of receptors in eukaryotic organisms and the target of ~30% of approved small molecule drugs. Upon receptor activation via ligand binding, these GPCRs signal through multiple downstream effectors such as G proteins and β–arrestins. Some ligands drive activation of G protein over β–arrestin dependent signaling, or vice versa ‐ a phenomenon known as biased agonism . It is unclear if biased agonism is a pharmacologic byproduct of GPCR complexity or an innate property of GPCRs. To investigate the mechanisms and implications of biased agonism in an endogenous system, we studied CXCR3, a chemokine receptor primarily present on T cells, which binds three endogenous ligands, CXCL9, CXCL10, and CXCL11. Despite its role in cancer, atherosclerosis, and other inflammatory disorders, no drugs target CXCR3. The purpose of this research is to investigate the impact of biased signaling at CXCR3 on downstream signaling pathways and demonstrate how this phenomenon drives highly specific cellular outputs. Methods We transfected HEK293N with CXCR3 and stimulated them with vehicle control or the endogenous ligands. We used mass spectrometry with an antibody‐free strategy involving high‐pressure, high‐resolution separation coupled with intelligent selection and multiplexing to assess changes in the phosphoproteome. Similarly, we treated activated CD8+ human T‐cells with the aforementioned ligands and used RNA‐seq to assess changes in the transcriptome. Results Our phosphoproteomic analyses (Figure 1) identified 19349 unique phosphorylation sites (phosphosites) on 5519 different proteins, of which 1532 phosphosites were differentially regulated between treatment groups. Similarly, our transcriptomic analyses (Figure 2) identified 48162 unique transcripts from 12405 genes, of which 887 transcripts were differentially regulated between treatment groups. Clustering analyses of these data demonstrate that the profiles of CXCL9 and CXCL10 differ greatly from that of CXCL11 which is concordant with published data demonstrating that CXCL9 and CXCL10 activate G protein and β–arrestin signaling in a balanced manner whereas CXCL11 is β–arrestin biased. Further analyses of these data demonstrate biased regulation of pathways regulating endocytosis, the cortical cytoskeleton, cell‐cell adhesion, cell division/cycle, ubiquitination, and the Jak‐STAT and MAPK signaling pathways. Conclusions We demonstrate the endogenous ligands of CXCR3 change the phosphoproteome and transcriptome in a biased manner. Our work demonstrates that, although these ligands signal through the same receptor, they are not redundant in function. Further exploration of the properties of these chemokines and CXCR3 may highlight the therapeutic promise of developing highly specific biased agonists to treat a variety of inflammatory disorders. Support or Funding Information Medical Scientist Training Program (DSE), T32GM00717144 (DSE), NIH (DSE, SR)Global Phosphoproteomic Analysis in HEK293 Cells overexpressing CXCR3Global Transcriptomic Analysis in Activated CD8+ Donor Human T‐Cells expressing CXCR3