NMR structure of CXCR3 binding chemokine CXCL11 (ITAC)
Abstract
CXCL11 (ITAC) is one of three chemokines known to bind the receptor CXCR3, the two others being CXCL9 (Mig) and CXCL10 (IP-10). CXCL11 differs from the other CXCR3 ligands in both the strength and the particularities of its receptor interactions: It has a higher affinity, is a stronger agonist, and behaves differently when critical N-terminal residues are deleted. The structure of CXCL11 was determined using solution NMR to allow comparison with that of CXCL10 and help elucidate the source of the differences. CXCL11 takes on the canonical chemokine fold but exhibits greater conformational flexibility than has been observed for related chemokines under the same sample conditions. Unlike related chemokines such as IP-10 and IL-8, ITAC does not appear to form dimers at millimolar concentrations. The origin for this behavior can be found in the solution structure, which indicates a β-bulge in β-strand 1 that distorts the dimerization interface used by other CXC chemokines.
ITAC (CXCL11) is a member of the chemokine family of small secretory proteins involved in immune and inflammatory responses. Chemokines play a key role in these processes by promoting recruitment and activation of different subpopulations of leukocytes (Mackay 2001). They fall into two main subfamilies based on the arrangement of the first two of four conserved cysteines, adjacent in CC chemokines and separated by one amino acid in CXC chemokines (Fernandez and Lolis 2002). Chemokines exert their effects by interacting with an appropriate seven-transmembrane G protein-coupled receptor on the surface of a leukocyte (Rojo et al. 1999). Knowledge of the structural basis for the interaction of chemokines with their receptors is of significant importance to understanding chemokine function and to the design of therapeutic interventions for a number of pathologies that involve chemokines (Power and Proudfoot 2001).
The receptors follow a similar nomenclature to the chemokines, with CC receptors generally interacting with CC chemokines and CXC receptors generally with CXC chemokines, although there are some examples of binding across families. ITAC binds to and activates the receptor CXCR3, as do IP-10 (CXCL10) and Mig (CXCL9; Loetscher et al. 1998). While ITAC, IP-10, and Mig are agonists for CXCR3, they can also act as antagonists for CCR3 (Loetscher et al. 2001). As well as binding to the same receptors, ITAC, IP-10, and Mig also show similarities to each other in that they share an individual branch of the phylogenetic tree (O’Donovan et al. 1999), are induced primarily by IFN-γ and are produced by macrophages as well as other cell types (Farber 1997; Cole et al. 1998; Laich et al. 1999; Tensen et al. 1999). However, ITAC also shows a number of functional differences from IP-10 and Mig.
ITAC has higher binding affinity for CXCR3 than either IP-10 or Mig, and is a more potent activator of CXCR3 (Meyer et al. 2001; Sauty et al. 2001; Clark-Lewis et al. 2003). ITAC is also a more potent antagonist to CCR3 than IP-10 or Mig (Loetscher et al. 2001). Structure–activity studies have shown that if its first three residues are deleted, ITAC retains significant binding affinity but loses the ability to activate CXCR3. By contrast, deletion of the N-terminal few residues of IP-10, Mig, or eotaxin (an agonist of CCR3) causes these proteins to lose binding capacity (Proost et al. 2001; Clark-Lewis et al. 2003). N-terminal deletions of ITAC are physiologically relevant, as the peptidase DPP-IV has been found to cleave the N-terminal two residues of ITAC, IP-10, and Mig in vivo, and thus regulates their effects on CXCR3 (Proost et al. 2001; Ludwig et al. 2002). Interestingly, the time scale observed for DPP-IV cleavage is significantly faster for ITAC than for IP-10 or Mig (Proost et al. 2001). We have previously determined the structure of IP-10 (Booth et al. 2002), and were interested in determining the structure of ITAC, in part to look for clues as to the origin of the observed differences in binding between the two proteins.
Acknowledgments
This work was funded by a grant from the PENCE. V.B. is supported by fellowships from the Canadian Institutes of Health Research (CIHR) and the Alberta Heritage Fund for Medical Research (AHFMR). We acknowledge the Canadian National High Field NMR Centre (NANUC) for their assistance and the use of the facilities. Operation of NANUC is funded by CIHR, the Natural Science and Engineering Research Council of Canada (NSERC), and the University of Alberta. Thanks also to Jan Rainey for help in preparing the CXCR3 receptor model, and Steffen Graether for comments on the manuscript.
The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 USC section 1734 solely to indicate this fact.
Notes
Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.04791404.