Diacylglycerol lipase alpha promotes hepatocellular carcinoma progression and induces lenvatinib resistance by enhancing YAP activity

Abstract

As an important hydrolytic enzyme that yields 2-AG and free fatty acids, diacylglycerol lipase alpha (DAGLA) is involved in exacerbating malignant phenotypes and cancer progression, but the role of the DAGLA/2-AG axis in HCC progression remains unclear. Here, we found that the upregulation of components of the DAGLA/2-AG axis in HCC samples is correlated with tumour stage and patient prognosis. In vitro and in vivo experiments demonstrated that the DAGLA/2-AG axis promoted HCC progression by regulating cell proliferation, invasion and metastasis. Mechanistically, the DAGLA/2AG axis significantly inhibited LATS1 and YAP phosphorylation, promoted YAP nuclear translocation and activity, and ultimately led to TEAD2 upregulation and increased PHLDA2 expression, which could be enhanced by DAGLA/2AG-induced activation of the PI3K/AKT pathway. More importantly, DAGLA induced resistance to lenvatinib therapy during HCC treatment. Our study demonstrates that inhibiting the DAGLA/2-AG axis could be a novel therapeutic strategy to inhibit HCC progression and enhance the therapeutic effects of TKIs, which warrant further clinical studies.

Fig. 1. DAGLA is upregulated in HCC samples and is correlated with poor prognosis in HCC patients.

A The difference in the 2-AG level between HCC tissues and adjacent nontumor tissues was detected by ELISA. B TCGA database showed that DAGLA mRNA was upregulated in a series of cancers, including HCC, compared with corresponding normal tissues. C TCGA and GTEx databases revealed that the DAGLA mRNA level was higher in tumour tissues than in normal liver tissues. D, E Kaplan–Meier analysis demonstrated that the DAGLA level was predictive of OS and RFS in HCC patients. F The proportion of stage III HCC patients in the high DAGLA group was markedly higher than that in the low DAGLA group. G qRT–PCR revealed that DAGLA was significantly upregulated in paired HCC tissue samples. H The change in DAGLA expression in representative HCC samples in a TMA by IHC staining. Scale bars, 100 μm. I DAGLA mRNA expression was positively correlated with the 2-AG concentration in HCC tissues. J, K Kaplan–Meier analysis of TMA patients indicated that high DAGLA expression in HCC tissues was correlated with worse OS and RFS.

Fig. 2. High DAGLA expression promotes HCC progression in vitro and in vivo.

A Western blotting confirmed the OE and KD efficiencies of DAGLA. B The 2-AG level in PLC/PRF/5-OE and Hep3B-KD cells was determined by ELISA. C–F CCK-8, EdU incorporation and colony formation assays were used to determine the role of DAGLA in HCC cell proliferation and colony formation. **P < 0.01. Scale bars, 100 μm. G Flow cytometric analysis demonstrated that Hep3B-KD induced G1/S arrest in Hep3B cells and that PLC/PRF/5-OE increased the proportion of PLC/PRF/5 cells in the S/G2 phase. H, I Transwell assays showed that DAGLA promoted the migration and invasion of HCC cells. J CCK-8 assays were used to verify the effect of exogenous 2-AG (5 µM) on Hep3B-KD cells. **P < 0.01. K Transwell assays were used to evaluate the effect of exogenous 2-AG on the migration and invasion abilities of Hep3B-KD cells. L, M Xenograft tumours derived from PLC/PRF/5-OE PLC/PRF/5 cells were significantly larger and tumours derived from Hep3B-KD Hep3B cells were significantly smaller than those derived from the corresponding control cells. N DAGLA promoted the lung metastasis of HCC cells in vivo. Scale bars, 100 μm.

Fig. 3. The tumorigenic role of DAGLA in HCC is mediated by the Hippo and PI3K/AKT signalling pathways.

A Heatmap of the DEGs in Hep3B-shDAGLA cells compared with control cells. B KEGG analysis of downregulated DEGs in Hep3B-shDAGLA cells revealed the potential signalling pathways regulated by DAGLA in HCC. C Heatmap of the representative genes whose expression was correlated with DAGLA expression identified by TCGA database analysis with the filtering criteria of a correlation coefficient ≥0.3 and a P-value < 0.05. D KEGG pathway enrichment analysis of the gene set positively correlated with DAGLA showed the enriched signalling pathways correlated with DAGLA in HCC. E Gene set enrichment analysis of Hep3B-KD and control cells. The normalised enrichment score (NES) is shown. F, G Western blotting analysis confirmed the effects of DAGLA OE and KD on the Hippo and PI3K/AKT signalling pathways in HCC. H, I A nuclear/cytoplasmic protein extraction assay revealed that Hep3B-shDAGLA and PLC/PRF/5-DAGLA increased and decreased the phosphorylation level of YAP in the cytoplasm, respectively. J The IF assay indicated that DAGLA drove the nuclear translocation of YAP in HCC cells. Scale bars, 10 μm. K Hep3B-shDAGLA cells were treated with exogenous 2-AG (5 µM) for 48 h, and WB analysis was performed to detect changes in the expression of the indicated proteins. L WB showed that exogenous 2-AG rescued the low nuclear YAP level resulting from DAGLA knockdown. M WB was performed to detect the expression of the indicated proteins. N IHC staining of the HCC TMA revealed the correlation between high DAGLA expression and YAP nuclear translocation. Scale bars, 150 μm.

Fig. 4. DAGLA facilitates HCC progression by regulating the Hippo and PI3K/AKT pathways.

A, B CCK-8 and Transwell assays were used to determine the effect of MK-2206 (10 μM, DMSO as a control) on PLC/PRF/5-DAGLA cells. **P < 0.01, ***P < 0.001. C Western blotting analysis revealed that MK-2206 treatment (48 h) increased the p-LATS1 and p-YAP levels in PLC/PRF/5-DAGLA cells to some extent. D IF staining and Western blotting analysis (E) showed that MK-2206 slightly blocked YAP nuclear translocation in PLC/PRF/5-DAGLA cells. Scale bars, 10 μm. F Western blotting analysis was used to determine the effect of SC79 (10 μM, 1 h; DMSO as the control) on the levels of the indicated proteins in Hep3B-shDAGLA cells. G WB demonstrated that SC79 facilitated YAP nuclear transport, which was partially inhibited by DAGLA downregulation. H qRT–PCR confirmed the efficiency of cotransfection of YAP and DAGLA modulation constructs. I–L CCK-8 and Transwell assays were used to explore the effects of YAP OE or KD on Hep3B-shDAGLA or PLC/PRF/5-DAGLA cells, respectively. **P < 0.01. M WB indicated the regulation of the Hippo signalling pathway in cell lines cotransfected with YAP and DAGLA modulation constructs.

Fig. 5. DAGLA promotes HCC progression by upregulating PHLDA2 expression induced by YAP nuclear translocation.

A The intersection of DEGs identified by TCGA database analysis, RNA-seq and DAGLA-correlated genes is shown in a Venn diagram. B, C TCGA database analysis indicated that patients with high PHLDA2 expression had worse OS and RFS. D, E The results of TMA IHC staining revealed that high PHLDA2 expression was associated with worse OS and RFS in HCC patients. F IHC staining of the HCC TMA demonstrated that high DAGLA levels were associated not only with higher YAP nuclear translocation but also with high PHLDA2 expression. Scale bars, 150 μm. G, H qRT–PCR and WB confirmed that DAGLA could regulate PHLDA2 expression in HCC cells. I WB showed the expression of PHLDA2 in Hep3B-shDAGLA+YAP-OE cells and PLC/PRF/5-DAGLA+YAP-KD cells. J WB showed that exogenous 2-AG rescued the low PHLDA2 expression induced by DAGLA knockdown. K ChIP‒qPCR revealed that TEAD2 bound to the PHLDA2 promoter in HCC cells. L–O CCK-8 and Transwell assays verified the biological effect of PHLDA2 in Hep3B-KD/OE cells. **P < 0.01.

Fig. 6. PHLDA2 promotes HCC progression by facilitating cell proliferation and inducing the EMT process.

A GSEA results of gene sets related to EMT, cell migration, liver cancer proliferation and cell cycle related in Hep3B-shDAGLA cells compared with the control group. B Heatmap of the expression of EMT- and cell migration-related genes in Hep3B-shDAGLA cells compared with the control group. C Heatmap of the expression of liver cancer proliferation- and cell cycle-related genes in Hep3B-shDAGLA cells compared with the control group. D The CDKN1C (p57) mRNA level was determined by qRT–PCR. E IF staining revealed the change in the E-cadherin levels in DAGLA-OE/KD cells. Scale bars, 10 μm. F Western blotting revealed the influence of DAGLA on the protein levels of key EMT mediators and p57. G Western blotting analysis clarified the effect of PHLDA2 on the expression of key EMT mediators and p57. H The levels of p57 and EMT-related proteins were verified by WB in PLC/PRF/5-DAGLA+PHLDA2-KD cells and Hep3B-shDAGLA+PHLDA2-OE cells. I Western blotting analysis showed the influence of exogenous 2-AG on EMT mediator expression in Hep3B-shDAGLA cells. J IHC staining demonstrated variations in the levels of p57, E-cadherin and vimentin in mouse xenograft tumours. Scale bars, 50 μm.

Fig. 7. DAGLA could modulate sensitivity to multikinase inhibitor therapy.

A–D CCK-8 and Transwell assays showed the effect of DAGLA OE and KD on the proliferation and invasion ability of HCC cells treated with lenvatinib (10 μM). E Experimental scheme. F–I Xenograft tumours derived from DAGLA-knockdown cells and treated with lenvatinib (10 mg/kg) showed better tumour growth inhibition, while tumours derived from DAGLA-overexpressing cells showed significant lenvatinib resistance, which could be reversed by verteporfin (40 mg/kg). J, K Comparison of OS curves between 36 advanced recurrent HCC patients with high or low DAGLA and PHLDA2 levels treated with lenvatinib and TACE who underwent liver resection before combined therapy. L Schematic depiction of the mechanisms underlying the DAGLA/2-AG axis regulating HCC progression and lenvatinib resistance by enhancing YAP nuclear translocation and activity.