Lipopolysaccharide promotes cancer cell migration and invasion through METTL3/PI3K/AKT signaling in human cholangiocarcinoma

Abstract

Purpose: As a major structural component of the outer membrane of Gram-negative bacteria, lipopolysaccharide (LPS) has been detected in the blood circulation and tissues in patients with chronic diseases and cancers, which plays a critical role in the tumor formation and progression. However, the biological role of LPS in human intrahepatic cholangiocarcinoma remains unclear. The aims of this study were to investigate the role of LPS in the malignant progression of intrahepatic cholangiocarcinoma.

Methods: The cell migration and invasion capacities of cholangiocarcinoma cell lines were evaluated by Boyden chamber assays. Expression levels of the key molecules involved in the PI3K/AKT signaling and METTL3 were detected by qPCR and western blot. The molecular mechanism by which LPS promotes the malignant behaviors was investigated by using siRNAs, plasmids and small molecule inhibitors.

Results: In vitro experiments showed that exogenous LPS treatment promoted cell migration and invasion capacities in both QBC939 and HUCCT1 cell lines, while did not affect cell proliferation and apoptosis. Mechanistically, exogenous LPS treatment had been proved to induce the increased expression of METTL3 and activate the downstream PI3K/AKTsignaling pathway. In addition, suppression of METTL3 expression reduced cell proliferation, migration and invasion capacities in both cell lines. Furthermore, inhibition of METTL3 expression or inhibition of PI3K/AKT signaling decreased LPS-induced cell migration and invasion capacities. Moreover, knockdown of METTL3 or inhibition of METTL3 significantly inhibited LPS-induced activation of the PI3K/AKT signaling.

Conclusion: In general, these results suggest that the LPS-METTL3-PI3K/AKT signal axis promotes cell migration and invasion in ICC, which contributes to a reduced overall survival in patients with ICC. It may broaden the horizon of cancer therapy with potential therapeutic targets.

Keywords: Cell invasion; Intrahepatic cholangiocarcinoma; Lipopolysaccharide; METTL3; PI3K/AKT signaling.

Fig. 1

Exogenous LPS treatment promotes cell migration and invasion while does not affect cell proliferation and apoptosis in human cholangiocarcinoma cells (A) QBC939 and HUCCT1 cells were treated with the indicated concentrations of LPS for 24 h and 48 h, cell growth was evaluated by CCK8 assay. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (B) QBC939 and HUCCT1 cells were treated with the low concentrations of LPS (1 μg/mL) for 24 h, cell apoptosis was detected by fluorescence activated cell sorting (FACS). Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (C) Cell migration ability and cell invasion ability of QBC939 and HUCCT1 cells treated by LPS (1 μg/mL) or PBS control. Representative images (left panel) and quantified analysis (right 2 panels) of transwell assays in QBC939 and HUCCT1 cells treated with and without LPS. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (D) QBC939 and HUCCT1 cells were treated with LPS (1 μg/mL) for 24 h and EMT related genes were detected by qPCR in both cell lines. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test).

Fig. 2

LPS promotes cell migration and invasion through activation of PI3K/AKT signaling pathway in human cholangiocarcinoma cells (A) Volcano plots of differential genes in QBC939 cells treated with LPS (1 μg/mL) for 24h (verse to control). (B) Results of KEGG enrichment analysis of differential genes. (C) QBC939 and HUCCT1 cells were treated with the indicated concentrations of LPS for 24 h, cell lysates were then analyzed for PI3K, p-PI3K, AKT and p-AKT expression by Western blot. GAPDH was used as loading control. Densitometric quantification is shown above the Western blots, representing the signal normalized to the loading control (GAPDH), relative to the untreated controls. (D)QBC939 and HUCCT1 cells were firstly incubated MK-2206 for 24 h, and then further treated with LPS (1 μg/mL) for another 24 h, cell lysates were then analyzed for AKT and p-AKT expression by Western blot. GAPDH was used as loading control. Densitometric quantification is shown above the Western blots, representing the signal normalized to the loading control (GAPDH), relative to the untreated controls. (E and F) QBC939 and HUCCT1 cells were firstly incubated MK-2206 for 24 h, and then further treated with LPS (1 μg/mL) for another 24 h, cell migration and invasion capacities were respectively evaluated by transwell migration and invasion assay. Representative images (left panel) and quantified analysis (right 2 panels) of transwell assays in QBC939 and HUCCT1 cells treated with and without LPS. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test).

Fig. 3

TLR4 is involved in LPS-induced cell migration in human cholangiocarcinoma cells (A) QBC939 and HUCCT1 cells were treated with LPS (1 μg/mL) for 24 h and TLR4 mRNA levels were detected by qPCR in both cell lines. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (B) QBC939 and HUCCT1 cells were transfected with three different TLR4 siRNA (siTLR4) and negative control for 24 h, RNA was extracted and analyzed for METTL3 mRNA expression by Real-Time Quantitative PCR. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (C and D). QBC939 and HUCCT1 cells were transfected with TLR4 siRNA (siTLR4) or scrambled siRNA (NC) for 24 h, then treated with LPS (1 μg/mL) for another 24 h, cell migration was evaluated by transwell migration assays. (E and F). QBC939 and HUCCT1 cells were transfected with TLR4 siRNA (siTLR4) or scrambled siRNA (NC) for 24 h, then treated with LPS (1 μg/mL) for another 24 h, cell invasion was evaluated by transwell invasion assays. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test).

Fig. 4

Exogenous LPS induces the expression of METTL3 which can regulate cell proliferation in human cholangiocarcinoma cells (A)QBC939 and HUCCT1 cells were treated with the indicated concentrations of LPS, transfected with METTL3 siRNA (siMETTL3) or METTL3 overexpression plasmid for 24 h, RNA was extracted and analyzed for METTL3 mRNA expression by Real-Time Quantitative PCR. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (B) QBC939 and HUCCT1 cells were treated with the indicated concentrations of LPS, transfected with METTL3 siRNA (siMETTL3) or METTL3 overexpression plasmid for 24 h, cell lysates were then analyzed for METTL3 expression by Western blot. β-actin was used as loading control. Densitometric quantification is shown above the Western blots, representing the signal normalized to the loading control (β-actin), relative to the untreated controls. (C) QBC939 and HUCCT1 cells were transfected with METTL3 siRNA (siMETTL3) or scrambled siRNA (siRNA) for 24 h, cell growth was evaluated by CCK8 assay. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test). (D) QBC939 and HUCCT1 cells were transfected with METTL3 overexpression plasmid or empty-vector for 24 h, cell growth was evaluated by CCK8 assay. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test).

Fig. 5

METTL3 is involved in LPS-promoted cell migration and invasion in human cholangiocarcinoma cells (A–D) QBC939 and HUCCT1 cells were transfected with METTL3 siRNA (siMETTL3) or scrambled siRNA (siRNA) for 24 h, then treated with LPS for another 24 h, cell migration and invasion were evaluated by transwell migration and invasion assays. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test).

Fig. 6

METTL3 regulates PI3K/AKT signaling pathway in human cholangiocarcinoma cells (A and B) QBC939 and HUCCT1 cells were transfected with METTL3 siRNA (siMETTL3) or scrambled siRNA (siRNA) for 24 h, then treated with LPS for another 24 h. Cell lysates were then analyzed for AKT and p-AKT expression by Western blot. GAPDH was used as loading control. (C and D) QBC939 and HUCCT1 cells were treated with indicated concentrations of STM2457 or DMSO for 24 h, then treated with LPS for another 24 h, cell lysates were then analyzed for AKT and p-AKT expression by Western blot. GAPDH was used as loading control. Densitometric quantification is shown next to the Western blots, representing the signal normalized to total AKT, relative to the untreated controls. Data represents means ± SD, n = 3. *p < 0.05, **p < 0.001, ***p < 0.001, n.s., non-significant (Student t-test).

Fig. 7

METTL3 and LPS upregulation predicts worse clinical outcomes in patients with ICC (A) The overall survival curve stratified by METTL3 mRNA expression levels in the ICC cohort (OEP001105). The p value was calculated using the log-rank test. (B) The overall survival curve stratified by LPS_mediated_signaling_pathway enrich score in the CCA cohort (OEP001105). The P value was calculated using the log-rank test. (C) The overall survival curve stratified by LPS_immune_ receptor_activity enrich score in the CCA cohort (OEP001105). The P value was calculated using the log-rank test.