CCR9-mediated signaling through β-catenin and identification of a novel CCR9 antagonist

Abstract

Elevated levels of chemokine receptor CCR9 expression in solid tumors may contribute to poor patient prognosis. In this study, we characterized a novel CCR9-mediated pathway that promotes pancreatic cancer cell invasion and drug resistance, indicating that CCR9 may play a critical role in cancer progression through activation of β-catenin. We noted that the CCL25/CCR9 axis in pancreatic cancer cells induced the activation of β-catenin, which enhanced cell proliferation, invasion, and drug resistance. CCR9-mediated activation of β-catenin and the resulting downstream effects were effectively inhibited by blockade of the PI3K/AKT pathway, but not by antagonism of Wnt. Importantly, we discovered that CCR9/CCL25 increased the lethal dose of gemcitabine, suggesting decreased efficacy of anti-cancer drugs with CCR9 signaling. Through in silico computational modeling, we identified candidate CCR9 antagonists and tested their effects on CCR9/β-catenin regulation of cell signaling and drug sensitivity. When combined with gemcitabine, it resulted in synergistic cytotoxicity. Our results show that CCR9/β-catenin signaling enhances pancreatic cancer invasiveness and chemoresistance, and may be a highly novel therapeutic target.

Keywords: CCL25; CCR9; Drug resistance; Pancreatic cancer; β-catenin.

Figure 1

Quantitative analysis of activated signaling pathways following CCL25 treatment. A) Phospho‐specific antibody microarrays were utilized to identify key signaling pathways regulated by CCL25 in MIAPaCa‐2 cells. The spot for active β‐catenin in the phospho‐protein array was marked with circle. B) Ratios of phosphorylated to unphosphorylated proteins quantified by fluorescent intensity in CCL25‐treated cells were calculated and compared to the untreated cell.

Figure 2

β‐catenin is activated following CCL25 exposure. A) PANC‐1, AsPC‐1 and MIAPaCa‐2 cells were treated with CCL25 for the indicated times and expression of activated β‐catenin was quantified after protein blotting. B) Increased expression and nuclear localization of active β‐catenin by CCL25 stimulation in MIAPaCa‐2 cells. C) Increased expression of active β‐catenin by CCL25 stimulation in the cell lysates extracted from both cytoplasmic and nuclear cell fractions. GAPDH and p84 were used as loading control for cytoplasmic and nuclear lysates, respectively.

Figure 3

CCL25‐mediated activation of β‐catenin is PI3K‐dependent, but Wnt‐independent. A) CCL25‐treated PANC‐1, AsPC‐1 and MiaPaCa2 cells for the indicated times induced activation of AKT (left), and treatment with LY294002 suppressed CCL25‐mediated p‐AKT expression (right). B) IWR1‐endo was able to block β‐catenin activation following Wnt‐3a stimulation, but did not inhibit CCL25‐mediated activation of β‐catenin (left). The β‐catenin‐specific inhibitor, CCT, blocked CCL25‐mediated activation of β‐catenin and inhibited expression of cyclin E (right). C) The activation of β‐catenin induced phosphorylation of pGSK‐3β. D) Protein expression levels of cyclin D1 (left) and E‐Cadherin (right) were also increased. Protein lysates were collected for immunoblotting 15 min after CCL25 treatment from pancreatic cancer cells.

Figure 4

CCL25 increases proliferation and invasion of pancreatic cancer cell lines and augments resistance to gemcitabine. A) PANC‐1, MIAPaCa‐2 and AsPC‐1 cells were treated with CCL25 under serum‐free condition for 72 h with and without LY294002 prior to assessing proliferation. B) PANC‐1 and MIAPaCa‐2 cells were plated in the Boyden chamber with or without LY294002. CCL25 was used as a chemoattractant in the serum free media, and cells were allowed to invade for 24 h. The invading cells were stained and counted at 5 different fields per experiment. The left bottom panels show photographs taken of a representative field in each condition. C) Pre‐treatment of PANC‐1, MIAPaCa‐2 and ASPC‐1 cells with CCL25 increased IC50 values for gemcitabine treatment.

Figure 5

Computational modeling of CCR9. A) Predicted binding site of antagonist CCX282 and the residues of interaction in the CCR9 structural model. B) Five potent CCR9 antagonist compounds screened from the NCI small molecular compound library. C) Predicted binding site of Compound 26 in CCR9. D) Compound 26 inhibited CCL25‐mediated activation of β‐catenin and AKT. Cells were pre‐treated with Compound 26 for 5 min prior to exposure to CCL25 for 15 min.

Figure 6

Compound 26 antagonizes CCL25 receptor mediated signaling. A) PANC‐1, MIAPaCa‐2 and ASPC‐1 cells were pre‐treated with Compound 26 (4uM) for 5 min prior to CCL25 treatment and cell growth was measured after 72 h. B) Pre‐treatment of Compound 26 inhibited CCL25‐mediated invasion in PANC‐1 cells. C) Synergistic effects on cell death were observed with the combination of Compound 26 and gemcitabine.