A role for chemokine receptor transactivation in growth factor signaling

Abstract

Complex cell responses require the integration of signals delivered through different pathways. We show that insulin-like growth factor (IGF)-I induces specific transactivation of the Gi-coupled chemokine receptor CCR5, triggering its tyrosine phosphorylation and Galpha recruitment. This transactivation occurs via a mechanism involving transcriptional upregulation and secretion of RANTES, the natural CCR5 ligand. CCR5 transactivation is an essential downstream signal in IGF-I-induced cell chemotaxis, as abrogation of CCR5 function with a transdominant-negative KDELccr5A32 mutant abolishes IGF-I-induced migration. The relevance of this transactivation pathway was shown in vivo, as KDELccr5A32 overexpression prevents invasion by highly metastatic tumor cells; conversely, RANTES overexpression confers built-in invasive capacity on a non-invasive tumor cell line. Our results suggest that this extracellular growth factor-chemokine network represents a general mechanism connecting tumorigenesis and inflammation.

Fig. 1. IGF-I induces specific CCR5 transactivation. (A) cells were untreated or Ptx-pretreated before assay for chemotaxis by the indicated stimuli. Ptx does not affect MCF-7 cell viability in these conditions (not shown). The results show the number of cells per field (200×) of Ptx-treated (open bar) or untreated (solid bar) MCF-7 cells stimulated by IGF-I or RANTES (n = 2; * p <0.01, two-tailed t-test). Dashed line indicates chemotaxis in the absence of stimulus. Serum-starved cells were stimulated with IGF-I, RANTES (B), or SDF-1α (C), and IGF-1R, CCR5 or CXCR4 selectively precipitated (i.p.). Filters were probed sequentially (w.b.) with anti-phosphotyrosine (PY) mAb, -IGF-1R, -CCR5, -Gα_i or anti-CXCR4 mAb.

Fig. 2. IGF-I-induced CCR5 transactivation is dependent on RANTES secretion. MCF-7 cells were stimulated with IGF-I alone or in the presence of anti-RANTES mAb (A), or preincubated with monensin before IGF-I stimulation (B). IGF-1R or CCR5 were selectively precipitated; Gα_i recruitment to CCR5, or IGF-1R autophosphorylation (PY) were analyzed by western blotting. Membranes were reprobed with anti-CCR5 and -IGF-1R antibodies. For monensin treatment, western blot data are represented as the densitometry intensity ratio for Gα_i and CCR5, or that calculated for tyrosine-phosphorylated and total IGF-1R β-subunit (** p <0.001, two-tailed t-test). (C) IGF-I-induced MCF-7 chemotaxis was assayed in the presence of neutralizing anti-chemokine mAb. An isotype-matched non-specific (non-sp) and a blocking anti-IGF-I mAb were included as negative and positive controls, respectively. Results are expressed as a percentage, and indicate inhibition of chemotaxis in the presence of each antibody (n = 3; ** p <0.001, two-tailed t-test). (D) RANTES levels of cell-free supernatants from serum-depleted (–) and IGF-I (10 nM; +)-stimulated cells.

Fig. 3. IGF-I regulates RANTES expression. (A) Equal amounts of lysates from unstimulated (–) or IGF-I-stimulated (+) cells were assayed for RANTES in ELISA (n = 3). (B) RANTES or GADPH mRNA were stimated in unstimulated (–) or stimulated with IGF-I alone (solid bars) or in the presence of actinomycin D (open bars). The graph represents the RANTES/GAPDH mRNA ratio expressed in arbitrary units (n = 3; ** p <0.001, two-tailed t-test). (C) MCF-7 cells transfected with a luciferase reporter gene under transcriptional control of RANTES (solid bars) or MIP-1β (open bars) promoters, unstimulated (–) or IGF-I-stimulated (+) for 5 or 24 h (** p <0.001, two-tailed t-test).

Fig. 4. Chemokine receptor signals are required for growth factor-induced cell polarization and chemotaxis. (A) CCR2, CCR5 and CXCR4 levels in mock- (dotted line) and KDELccr5Δ32- (solid line) transduced cells were estimated by flow cytometry. Background fluorescence of an isotype-matched non-specific mAb is shown in gray. (B) IGF-I (solid bars), IGF-I+RANTES (open bars) or IGF-I+SDF-1α (gray bars)-induced chemotaxis of mock- and KDELccr5Δ32-transduced cells. The number of migrating mock-transduced cells using IGF-I as chemoattractant was considered 100% (n = 3). (C) Paxillin staining (red) of mock- and KDELccr5Δ32-expressing MCF-7 cells untreated (none), or stimulated with IGF-I, RANTES or SDF-1α; transduced cells co-express GFP (green). Bar, 10 µm. (D) Quantification of random fields from (C) (three independent experiments; ** p <0.001, two-tailed t-test).

Fig. 5. Chemokine receptor signals modulate tumor invasiveness in vivo. (A) RANTES levels measured by ELISA in cell-free supernatants from DU-145. (B) Mock and KDELccr5Δ32-DU-145 (10⁶ cells) were inoculated onto chick embryo CAM, and intravasated cells estimated (three independent experiments, n = 16 for each cell line). (C) Mock- and RANTES-MCF-7 (2 x 10⁶ cells) were inoculated onto the CAM as in (B). Three independent experiments, n = 14 for each cell line.