CD34+ hematopoietic stem- and progenitor cells (HPC) express high levels of the G protein-coupled receptors (GPCR) CXCR4 and CysLT1. In contrast to the established role of CXCR4 in stem cell homing, the function of CysLT1 in HPC remains only partially understood. We found that upon stimulation of peripheral blood CD34+ HPC in vitro with the respective ligands (CXCL12/SDF1 for CXCR4 and the cysteinyl-leukotriene LTD4 for CysLT1), both receptors similarly mediate the following functional activities: intracellular calcium fluxes, actin polymerization, adhesion to endothelium in vitro and chemotaxis. By Westernblot analysis, we could demonstrate Pyk2- and MAP-kinase phosphorylation as pivotal elements of CysLT1 signaling. These pathways have previously been identified also in CXCR4 signaling.
To further analyze signal transduction pathways of both receptors, CD34+ cells were pretreated with pertussis toxin (PTX) or with a specific PKC inhibitor, which is an isoform-specific inhibitory myristoylated peptide derived from the pseudosubstrate (PS) region of PKCzeta, mimics the substrate, and maintains PKC in its nonactive isoform. Subsequently, we determined actin polymerization by flow cytometry using phalloidin-FITC, and adhesion to IL-1 stimulated endothelial cells (HUVEC). CXCR4 signaling leading to actin polymerization was found to be completely blocked by preincubation with pertussis toxin (PTX) and therefore is mediated exclusively by Gi proteins, while CysLT1 also involves Gq proteins as reflected by only partial inhibition by PTX. For both receptors, the Pyk2 signaling pathway leading to actin polymerization and adhesion was completely suppressed by preincubation with PSzeta and therefore dependent on atypical PKCzeta, which is calcium and DAG independent.
We further examined whether these two similarly functioning receptors maintain any crosstalk, as has been reported for GPCR. Their possible interaction was explored using actin polymerization as a functional read-out. CXCR4- and CysLT1-mediated actin polymerization in response to their respective ligand was induced within 10 sec and returned to basal levels after 4 min. A second challenge after 4 min with the same ligand resulted in a complete suppression, demonstrating self-desensitization of both receptors. Interestingly, restimulating CXCL12-induced cells with LTD4 after 4 min resulted in complete suppression of actin polymerization, whereas restimulating LTD4-induced cells with CXCL12 lead to F-actin levels comparable to those achieved with the first challenge. These data show for the first time that CXCR4 can cross-desensitize CysLT1 while there is no crosstalk from CysLT1 to CXCR4.
In conclusion, CXCR4 and CysLT1 share major signaling pathways. However, the ability to desensitize other GPCR is strikingly different. The finding that CXCR4 cross-desensitizes CysLT1 but not vice versa could explain our observation that a CysLT1 antagonist (montelukast) did not mobilize HPC in vivo, as the presence of CXCL12 in the stem cell niche may result in desensitization of CysLT1. In contrast, CXCR4-dependent bone marrow homing may not be influenced by conditions with high local and systemic cysteinyl-leukotriene concentrations, e.g. during allergy and inflammation. In the absence of CXCR4 activation however, CysLT1 could be important for homing of stem and progenitor cells in areas other than the bone marrow with a high local concentration of cysteinyl leukotrienes, e.g. in inflamed tissues.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.