The CXC-motiv-chemokine receptor 7 (CXCR7), also known as atypical chemokine receptor 3 (ACKR3), is a non classical seven transmembrane-spanning receptor for chemokines CXCL11 and -12 and migration inhibitory factor (MIF). CXCR7 is involved in a variety of diseases including cardiovascular diseases, thrombosis, and thrombo-inflammation. CXCR7 has become an attractive target to control thrombosis and organ ischemia in coronary artery disease.

To identify novel chemical structures for specific modulation of CXCR7 function, we developed a 3D model of CXCR7 based on protein structure of similar crystallized receptors and potential binding mode and performed in silico screening of a large virtual compound library for potential CXCR7 directed compounds. Approximately 1400 hits were identified and 35 potential candidates were tested in biological platelet function assays. The in vitro results and the chemical structure of 6 highly reactive candidates were used to refine the 3D model of CXCR7 followed by a second round of in silico screening of virtual compound library. After testing of 47 identified potential  CXCR7 chemical structures using the ß-arrestin Pathhunter^R assay (eurofins DiscoverX) we selected a lead compound (C10) for further biological assessment. C10 inhibited agonist-dependent (ADP, CRP) degranulation of P-selectin (flow cytometry) in a concentration dependent manner in contrast to a control C46 molecule with similar molecular weight. Further, agonist-induced platelet apoptosis (loss of mitochondrial potential) was significantly  reduced in the presence of C10. Flow-mediated platelet-dependent thrombus formation on immobilized collagen was also substantially decreased by approximately 30% in the presence of C10 compared to C46 control. To test for CXCR7 specificity we made use of a knock-out mouse strain with megakaryocyte/platelet-specific genetic deletion of CXCR7. Compared to wild type littermates (CXCR7^+/+), platelets deficient in CXCR7 (CXCR7^-/-) did not show a reduced platelet activation and degranulation, aggregation, or platelet-dependent thrombus formation under flow. This indicates a highly specific activity of the identified lead compound C10 for CXCR7.

The present work shows, that  a computerized  generation of a 3D model  of the chemokine receptor CXCR7 and subsequent in silico screening of a large virtual compound library is a realistic and efficient tool to identify lead compounds for anti-thrombotic activity. Targeting CXCR7 may be a future possibility to modulate thrombosis without enhancing hemostasis and thus bleeding.