Chemokines and their receptors: insights from molecular modeling and crystallography

Abstract

Chemokines are small secreted proteins that direct cell migration in development, immunity, inflammation, and cancer. They do so by binding and activating specific G protein coupled receptors on the surface of migrating cells. Despite the importance of receptor:chemokine interactions, their structural basis remained unclear for a long time. In 2015, the first atomic resolution insights were obtained with the publication of X-ray structures for two distantly related receptors bound to chemokines. In conjunction with experiment-guided molecular modeling, the structures suggest a conserved receptor:chemokine complex architecture, while highlighting the diverse details and functional roles of individual interaction epitopes. Novel findings promote the development and detailed structural interpretation of the canonical two-site hypothesis of receptor:chemokine recognition, and suggest new avenues for pharmacological modulation of chemokine receptors.

Figure 1. Chemokines and their preferred interaction interfaces

(A–D) Ribbon diagrams of four families of chemokines found in mammals: CC, CXC, CX₃C, and XC. The different spacing of the conserved N-terminal cysteines causes variations in the conformations of the flexible N-termini, marked by arrows. Ribbon coloring is the same as in (I–L). (E–H) Preferred dimerization geometry of chemokines from different families. (I–L) Representative chemokines from each family are shown as molecular surfaces and colored according to the frequency and strength of inter-molecular contacts that they make with diverse binding partners in the available crystal structures: darker red color indicates a preferred interaction interface.

Figure 2. Structural insights into receptor:chemokine recognition

(A) The canonical two-site hypothesis of receptor:chemokine interaction: the flexible sulfotyrosinated N-terminus of the receptor (chemokine recognition site 1, CRS1, important for binding affinity) binds to the globular core of the chemokine, while the transmembrane (TM) domain pocket of the receptor (chemokine recognition site 2, CRS2, critical for signaling) accommodates the distal N-terminus of the chemokine. (B–D) NMR structures of chemokines with isolated N-termini of their receptors represent CRS1 interactions outside the full length receptor context. The structure of CXCL8 with a chemically modified N-terminus of CXCR1 [49] (B) is the only structure where the receptor peptide orientation is consistent with the full-length receptor complex. (E) An X-ray structure of CXCL12 in complex with a small molecule inhibitor and a sulfate ion: the inhibitor utilizes the preferred interaction surface of the chemokine. (F–G) The first two X-ray structures of full-length receptor:chemokine complexes: the presence of receptor TM domains imposes constraints onto the geometry of CRS1 interactions. (H) CXCR4 undergoes a large conformational change in order to accommodate a chemokine.

Figure 3. Determinants of CC vs CXC receptor:chemokine specificity elucidated by crystallography in conjunction with molecular modeling and bioinformatics

Potential specificity-determining positions are highlighted in the partial sequence alignments and mapped onto structures and models for human CC chemokines (A), CC receptors (B), CXC chemokines (C), and CXC receptors(D). The highlighted chemokine and receptor features are complementary and are predicted by modeling to directly interact in complexes that belong to the respective classes.

Figure 4. Receptor:chemokine interaction epitopes

(A) Modeling explains the critical role of chemokine recognition site 2 (CRS2, green) in CXCR4:CXCL12 signaling. (B) The emerging architecture of receptor:chemokine complexes features at least five distinct interaction epitopes, possibly with diverse roles in binding affinity, signaling, and pharmacology. Some of the epitopes are yet unnamed (labeled “CRS?”).