CXCR7 Functions Together with CXCR4 in SDF-1/CXCL12 Induced CD34⁺ Hematopoietic Progenitor Cell Cycling Through β-Arrestin 2-Dependent Akt Activation

SDF-1/CXCL12 chemokine exhibits a well-known effect on retention, migration and homing of hematopoietic stem/progenitor cells (HSC/HP). We have previously demonstrated that it is also a key regulator of hematopoiesis homeostasis, acting, at low concentrations, as a survival and cell cycle promoting factor for human CD34⁺ HP. It has long been considered that CXCR4 was responsible for SDF-1 induced biological effects until the recent discovery of its second receptor, CXCR7. In the present study, we explored the respective role of CXCR4 and CXCR7 in the cell cycling and survival promoting effect of SDF-1/CXCL12 on human CD34⁺HP.

We used CD34⁺ HP purified from the peripheral blood (PB) of healthy un-mobilized donors since they are mainly in G₀. This allows to study the role of CXCR4 and CXCR7 receptors in 0.5ng/ml SDF-1/CXCL12 induced G₀-G₁ transition in synchronized quiescent cells. Gene expression was detected by RT-QPCR. Protein expression was detected and quantified using confocal microscopy, flow cytometry, immunoblotting and immunoprecipitation. Cell cycling experiments were performed using a Ki67 antibody and CXCR7 binding assay was performed using SDF-1/CXCL12^AF647. Neutralization experiments were performed using a specific CXCR4 antibody or CXCR7 chemical inhibitors, a kind gift from ChemoCentryx, Inc (CCX771 and CCX733) and their respective controls (IgG and CCX704).

Flow cytometry and confocal analysis showed that CXCR7 and CXCR4 are differentially distributed in PB CD34⁺ cells. In contrast to CXCR4 that is present at both the plasma membrane and intracellular level, CXCR7 expression is mainly restricted to the intracellular compartment. Confocal analysis suggested the presence of CXCR4/CXCR7 heterodimers on these cells the presence of which were confirmed by immunoprecipitation in a HP cell line. Despite its very low expression at the surface of CD34⁺ cells, we found that CXCR7 is capable of binding to exogenous SDF-1/CXCL12. Indeed, pretreatment with CXCR7 antibody or a chemical inhibitor reduces the mean fluorescence of bound fluorescent SDF-1/CXCL12^AF647, a fully functional and specific chemokine with similar effects compared to unlabeled SDF-1/CXCL12.

Neutralizing either CXCR4 or CXCR7 in PB CD34⁺ cells strongly reduced Akt activation induced by SDF-1/CXCL12 (0.5 ng/ml) as well as the percentage of cells in cycle (G₁ and S + G₂/M), colony formation and cell survival. This demonstrates that both receptors cooperate in SDF-1/CXCL12 induced functional effects.

We further analyzed the respective role of CXCR4 and CXCR7 in SDF-1/CXCL12 signalization. In contrast to CXCR4, CXCR7 is reported not to activate G protein signaling pathways in response to SDF-1/CXCL12. However, it can transduce cell signaling through the β-arrestin pathway. In the present study, we showed that CXCR7 and β-arrestin 2 colocalize near the plasma membrane in freshly purified PB CD34⁺ cells, suggesting that CXCR7 is constitutively activated. After SDF-1/CXCL12 treatment, the majority of β-arrestin 2 was translocated to the nucleus and only a partial colocalization persisted in the cytoplasm. Using neutralizing antibodies and specific inhibitors, we showed that β-arrestin 2 nuclear translocation was dependent on both CXCR7 and CXCR4 receptors. Reducing β-arrestin 2 expression using siRNA decreased SDF-1/CXCL12 induced Akt activation in PB CD34⁺cells indicating the involvement of β-arrestin 2 in this process.

Altogether, our results demonstrate for the first time the role of CXCR7 together with CXCR4 in SDF-1/CXCL12-induced CD34⁺ cell cycling/proliferation. They also suggest the involvement of β-arrestin 2 as signalling hubs, downstream of both receptors.

No relevant conflicts of interest to declare.