Fig 2: Berberine exhibited anti-tumor effect in orthotopic pancreatic cancer implantation model. (A) Oral administration of berberine had minimal effect on the body weight of C57BL/6N mice. (B) Compared to the control (CTL, n = 16) group, berberine (5 mg/kg, n = 16 and 10 mg/kg, n = 16) can prolong survival in pancreatic cancer model (P = 0.0294). (C) Representative images of the dissected pancreas and liver at the end of berberine treatment showing that berberine intervention can regress orthotopic growth of implanted pancreatic cancer and suppress liver metastasis. The graph represents the orthotopic tumor size of the controlled group mice (CTL, n = 6), 5 mg/kg berberine group (n = 9) and 10 mg/kg berberine group (n = 7) by the end of the 4-week intervention (**P < 0.01). (D) Compared to the control group, berberine treatment (5 mg/kg, n = 9 and10 mg/kg, n = 7) suppresses serum LDH level (*P < 0.05). (E) a shift of serum LDH isoenzyme to LDH-5 was found and berberine treatment group exhibit lower serum LDH-5 than that of the control (*P < 0.05). (F) Compared to the control (CTL, n = 6) group, berberine (5 mg/kg, n = 6, and 10 mg/kg, n = 6) treatment suppresses serum L-lactate level (*P < 0.05, **P < 0.01) in murine in orthotopic pancreatic cancer implantation model. (G) Compared to the control (CTL) group, berberine (5 mg/kg and 10 mg/kg) treatment suppresses intratumoral L-lactate level (n = 6 each group, **P < 0.01) in murine in orthotopic pancreatic cancer implantation model. (H) Representative histological and immunohistochemistry (IHC) photomicrographs of the tumor in individual groups. Paraffin-embedded tumor tissues were stained with hematoxylin and eosin (H&E), anti-Ki67 antibody and anti-LDHA antibody to determine the pathological type, proliferation, and LDHA expression of the tumor cells upon different treatments. The graph represents the percentage of Ki-67 and LDHA-positive cells of pancreatic and liver lesions in control group (CTL, n = 6), 5 mg/kg berberine group (n = 6) and 10 mg/kg berberine group (n = 6) (**P < 0.01, ***P < 0.001). (I) Representative histological and immunofluorescent (IF) photomicrographs of the pancreatic orthotopic lesions. Paraffin-embedded tumor tissues were stained with DAPI (Blue), anti-CD3 (Red), anti-CD4 (Green), and anti-CD8 (Green) antibody in individual groups upon different treatments. (J) The schematic representative of regulatory mechanism underlying LDHA-mediated inhibition of pancreatic cancer by berberine. CTL: control group; LDH: lactate dehydrogenase; HE: hematoxylin and eosin staining; LDHA: lactate dehydrogenase A

Fig 3: Berberine suppressed pancreatic cancer proliferation and invasion by LDHA/AMPK pathway. (A) Berberine may direct bind to LDHA, as predicted by molecular docking. (B) Surface plasmon resonance (SPR) biosensor analysis of berberine binding to LDHA. The sonograms for binding of berberine in serial concentration (1.5625, 3.125, 6.25, 12.5, 25 µM) to LDHA are shown. (C) Intracellular and extracellular lactate assay were performed with cells incubated with 5 or 10 µM of berberine for 48 hours. The graph represents the relative intensity of lactate detected intracellularly and in the culture supernatant normalized per cell number compared to the respective controlled (NC) cells (n = 3 per group, *P < 0.05, **P < 0.01). (D) Control Panc02 and Panc-1 cells were incubated with 5 or 10 µM of berberine (BBR) for 48 hours. The graph represents the relative protein expression ration of pAMPKa/AMPKa and LDHA normalized to beta-actin compared to the nontreated control cells (n = 3 per group, *P < 0.05, **P < 0.01, ***P < 0.001). (E) MTT assay showed that berberine suppressed Panc02 and Panc-1 pancreatic cancer cells proliferation and the cytotoxicities of berberine varies among different LDHA expression cells after 48 hours of incubation with berberine (n = 6 per group). (F) Colony formation assay was performed with controlled Panc02 and Panc-1 cells treated with 5 µM berberine with or without supplementation with 10 mM L-lactate. The graph represents the percentage of area covered by colonies after 10 days of incubation (n = 6 per group, **P < 0.01). (G) Boyden chamber invasion assay was performed for 24 hours with controlled Panc02 and Panc-1cells treated with 5 µM berberine with or without supplementation with 10 mM L-lactate. The graph represents the percentage of area covered by migrated cells after incubation (n = 6 per group, **P < 0.01). (H) Controlled Panc02 and Panc-1 cells were incubated with or without 5 µM of berberine and supplanted with or without 10 mM L-lactate for 24 hours. The graph represents the relative protein expression ration of pAMPKa/AMPKa and LDHA normalized to beta-actin compared to the nontreated control cells (n = 3 per group, *P < 0.05, **P < 0.01, ***P < 0.001). BBR: berberine; IC50: half maximal inhibitory concentration; NC: controlled cells; SH: LDHA knockdown cells; OE: LDHA overexpressed cells

Fig 4: L-lactate restores AMPK activation by LDHA knockdown and LDHA is an upstream of the AMPK-mTOR pathway. (A) The expression of the indicated proteins was analyzed by immunoblotting. The graph represents the relative protein expression ratio of pAMPKa/AMPKa and pmTOR/mTOR normalized to beta-actin compared to the controlled (NC) cells (n = 3 per group, **P < 0.01, ***P < 0.001). (B) LDHA knockdown (SH) cells were incubated with 10 mM L-lactate for 24 hours. The graph represents the relative protein expression ration of pAMPKa/AMPKa and pmTOR/mTOR normalized to beta-actin compared to the nontreated knockdown (SH) cells (n = 3 per group, *P < 0.05, ***P < 0.001). (C) LDHA knockdown (SH) cells were incubated with 20 µM AMPK inhibitor (Compound C) for 24 hours. The graph represents the relative protein expression ration of pAMPKa/AMPKa and pmTOR/mTOR normalized to beta-actin compared to the nontreated knockdown (SH) cells (n = 3 per group, *P < 0.05, **P < 0.01). (D) Colony formation assay was performed with nontreated LDHA knockdown (SH) cells and LDHA-SH cells supplement with 20 µM Compound C. The graph represents the percentage of area covered by colonies after 10 days of incubation (n = 6 per group, ***P < 0.001). (E) Boyden chamber invasion assay was performed for 24 hours with LDHA knockdown (SH) cells with or without 20 µM Compound C. The graph represents the percentage of area covered by migrated cells after incubation (n = 6 per group, **P < 0.01, ***P < 0.001). (F) LDHA overexpressed cells were incubated with 10 nm Rapamycin for 24 hours. The graph represents the relative protein expression ratio of pAMPKa/AMPKa and pmTOR/mTOR normalized to beta-actin compared to the nontreated overexpressed (OE) cells (n = 3 per group, *P < 0.05). (G) Colony formation assay was performed with LDHA overexpressed (OE) cells supplement with or without 10 nm Rapamycin. The graph represents the percentage of area covered by colonies after 10 days of incubation compared to the nontreated overexpressed (OE) cells (n = 6 per group, *P < 0.05). (H) Boyden chamber invasion assay was performed for 24 hours with LDHA overexpressed cells treated with or without 10 nm Rapamycin. The graph represents the percentage of area covered by migrated cells after incubation (n = 6 per group, ***P < 0.001). NC: controlled cells; SH: LDHA knockdown cells; OE: LDHA overexpressed cells

Fig 5: Exogenous overexpress LDHA enhances proliferation, migration, and invasion in vitro and promotes tumorigenesis and liver metastasis in vivo. (A) The protein expression of LDHA of various pancreatic cancer cell lines was confirmed by immunoblotting. Beta-actin was used as loading control. The mRNA expression of LDHA and invasion capability of various pancreatic cancer cell lines. Invasion capability through Matrigel and ordered from strongest to lowest, from Panc02 cells to Capan-2 cells (n = 3 per group). (B) The overexpression and knockdown of LDHA-transfected Panc02 and Panc-1 cells was confirmed by immunoblotting and RT-PCR. Beta-actin was used as loading control (n = 3 per group). (C) LDHA overexpression (OE) promoted the colony formation capabilities in Panc02 and Panc-1 cells (n = 6 per group, **P < 0.01, ***P < 0.001) compared with controlled cells (NC). LDHA knockdown (SH) suppressed colony formation in Panc-1 and Panc02 cells (n = 6 per group, *P < 0.05, **P < 0.01) compared to controlled cells (NC). (D) The invasive properties were analyzed by Matrigel-coated Boyden chamber assay and scored under a light microscope (200×). LDHA knockdown (SH) suppressed the invasive properties in Panc02 cells (**P < 0.01) and Panc-1 cells (*P < 0.05) compared to the respective controlled cells (NC) (n = 6 per group). Overexpression of LDHA (OE) promoted cell invasion in both Panc02 and Panc-1 cells (n = 6 per group, ***P < 0.01) compared to the respective controlled cells (NC). (E) A wound was produced and monitored at 0, 24, and 48 h as the cells moved and filled the damaged area (200×). The data were plotted as the percentage area healed (n = 3 per group, *P < 0.05, **P < 0.01). (F) The tumor size and average weight of primary xenografts of orthotopic implantation model (n = 6 per group, **P < 0.01). H&E staining confirmed the tumorigenesis of LDHA-OE cells in orthotopic model (400×). The Ki-67 and LDHA expression in the histological sections was detected by immunohistochemical (IHC) staining (n = 12 per group, ***P < 0.001, 400×). (G) Macroscopic liver metastasis lesion was visible on the surface of the liver tissue (white arrowed) in LDHA-OE group and the number of liver metastatic lesions in mice (n = 6 per group) was analyzed by counting the macroscopic lesion from each hepatic lobe (***P < 0.001). H&E staining identified adenocarcinoma in the liver section of the LDHA-OE group, while no liver metastasis lesion was identified with the control group (400×). The Ki-67 and LDHA expression in the histological sections was detected by immunohistochemical (IHC) staining (n = 12 per group, 400×). NC: controlled cells; SH: LDHA knockdown cells; OE: LDHA overexpressed cells