Human cholangiocarcinoma development is associated with dysregulation of opioidergic modulation of cholangiocyte growth

Abstract

Background/aims: Incidence of cholangiocarcinoma is increasing worldwide, yet remaining highly aggressive and with poor prognosis. The mechanisms that drive cholangiocyte transition towards malignant phenotype are obscure. Cholangiocyte benign proliferation is subjected to a self-limiting mechanism based on the autocrine release of endogenous opioid peptides. Despite the presence of both, ligands interact with delta opioid receptor (OR), but not with microOR, with the consequent inhibition of cell growth. We aimed to verify whether cholangiocarcinoma growth is associated with failure of opioidergic regulation of growth control.

Methods: We evaluated the effects of OR selective agonists on cholangiocarcinoma cell proliferation, migration and apoptosis. Intracellular signals were also characterised.

Results: Activation of microOR, but not deltaOR, increases cholangiocarcinoma cell growth. Such an effect is mediated by ERK1/2, PI3K and Ca(2+)-CamKIIalpha cascades, but not by cAMP/PKA and PKCalpha. microOR activation also enhances cholangiocarcinoma cell migration and reduces death by apoptosis. The anti-apoptotic effect of microOR was PI3K dependent.

Conclusions: Our data indicate that cholangiocarcinoma growth is associated with altered opioidergic regulation of cholangiocyte biology, thus opening new scenarios for future surveillance or early diagnostic strategies for cholangiocarcinoma.

Fig. 1

Evaluation of OR expression in cells of three different cholangiocarcinoma cell lines (Mz-ChA-1, TFK-1 and HuH-28) and in a hepatocarcinoma cell line (Alex-0 cells), assessed by immunoblotting. The cells of the three cholangiocarcinoma cell lines express both δOR and µOR, whereas no expression was found in hepatocarcinoma cells.

Fig. 2

Effect of OR activation on human cholangiocarcinoma cell proliferation cultured in FBS-free medium, measured by BrDU incorporation (A–C). Increasing doses of µOR (DAMGO), but not of δOR (DPDPE), selective agonist markedly increased HuH-28 (A), Mz-ChA-1 (B) and TFK-1 (C) cholangiocarcinoma cell proliferation. Data are expressed as mean ± S.E. of three experiments. *p < 0.05 vs. basal. (D) Effect of δOR activation on HuH-28 cell proliferation assessed by PCNA protein expression. Increasing concentrations of DPDPE did not determine changes in PCNA protein expression, neither in cells cultured in the presence (left) nor in the absence (right) of 10% FBS-enriched medium. Data are expressed as mean ± S.E. of three experiments. *p ≤ 0.05 vs. basal.

Fig. 3

Effect of µOR activation on intracellular signalling in HuH-28 cells. (A) Incubation with increasing doses of DAMGO corresponded to a dose-dependent increase of ERK1/2 (left) and AKT (right) phosphorylation. Data are expressed as mean ± S.E. of three experiments. ^#p < 0.05 vs. pERK1 basal value; *p < 0.05 vs. pERK2 basal value; ^◦p < 0.05 vs. pAKT basal value. (B) Incubation with increasing doses of DAMGO did not induce significant changes in PKA activity (left); active PKA provided by the vendor and cholangiocytes isolated from rats subjected to 1 week BDL were taken as positive controls (right, representative image). Data are expressed as mean ± S.E. of three experiments. (C) µOR activation corresponded to a dose-dependent increase of CamKIIα (left) but not PKCα (right) phosphorylation. Data are expressed as mean ± S.E. of three experiments. ^§p < 0.05 vs. pCamKIIα basal value.

Fig. 3

Effect of µOR activation on intracellular signalling in HuH-28 cells. (A) Incubation with increasing doses of DAMGO corresponded to a dose-dependent increase of ERK1/2 (left) and AKT (right) phosphorylation. Data are expressed as mean ± S.E. of three experiments. ^#p < 0.05 vs. pERK1 basal value; *p < 0.05 vs. pERK2 basal value; ^◦p < 0.05 vs. pAKT basal value. (B) Incubation with increasing doses of DAMGO did not induce significant changes in PKA activity (left); active PKA provided by the vendor and cholangiocytes isolated from rats subjected to 1 week BDL were taken as positive controls (right, representative image). Data are expressed as mean ± S.E. of three experiments. (C) µOR activation corresponded to a dose-dependent increase of CamKIIα (left) but not PKCα (right) phosphorylation. Data are expressed as mean ± S.E. of three experiments. ^§p < 0.05 vs. pCamKIIα basal value.

Fig. 4

(A) The DAMGO-induced increase in HuH-28 cell proliferation was neutralised by the pre-incubation with PI3K (wortmannin), MEK (PD98059) and CamKII (KN62) inhibitors and with the intracellular Ca²⁺ chelator (BAPTA/AM). In contrast, DAMGO-induced increase in cell proliferation was not affected by the pre-incubation with the cAMP-dependent PKA inhibitor (Rp-cAMPs) or the Ca²⁺-dependent PKC inhibitor (Ro-32-0432). Data are expressed as mean ± S.E. of three experiments. *p < 0.05 vs. the other groups. (B) Only PI3K and MEK inhibitors blocked the DAMGO-induced increase in ERK1/2 phosphorylation, whereas no effect was observed after the pre-incubation with the other inhibitors. Data are expressed as mean ± S.E. of three experiments. ^#p < 0.05 vs. pERK1 value of the other groups; ^§p < 0.05 vs. pERK2 value of the other groups. (C) DAMGO-induced increase in AKT phosphorylation was neutralised only by the PI3K inhibitor wortmannin. Data are expressed as mean ± S.E. of three experiments. ^◦p < 0.05 vs. the other groups.

Fig. 5

Effect of µOR activation on HuH-28 cell migration and survival. (A) Wound-healing experiments showed that µOR activation markedly diminished both after 24 and 72 h, the area included within the wound margins (outlined by the red-scattered line). Data are expressed as mean ± S.E. of three experiments. *p < 0.05 vs. the corresponding basal value. (B) Cell migration assay demonstrated that the µOR selective agonist DAMGO markedly enhanced HuH-28 cell migration; cells cultured in 10% FBS enriched medium were used as positive control. Data are expressed as mean ± S.E. of three experiments. ^◦p < 0.05 vs. basal.

Fig. 6

(A) Incubation of HuH-28 cells with GCDCA resulted in marked activation of caspase 3; such effect was reduced when cells were pre-incubated with DAMGO. Data are expressed as mean ± S.E. of three experiments. ^#p < 0.05 vs. basal; ^§p < 0.05 vs. GCDCA+DAMGO. (B) Only the pre-incubation with wortmannin neutralised the effects of DAMGO on GCDCA-induced increase in caspase 3 activity assay. No effects were seen when cells were pre-incubated with PD98059 or KN62. Data are expressed as mean ± S.E. of three experiments. *p < 0.05 vs. the other groups.