Targeting glucosylceramide synthase upregulation reverts sorafenib resistance in experimental hepatocellular carcinoma

Evasive mechanisms triggered by the tyrosine kinase inhibitor sorafenib reduce its efficacy in hepatocellular carcinoma (HCC) treatment. Drug-resistant cancer cells frequently exhibit sphingolipid dysregulation, reducing chemotherapeutic cytotoxicity via the induction of ceramide-degrading enzymes. However, the role of ceramide in sorafenib therapy and resistance in HCC has not been clearly established. Our data reveals that ceramide-modifying enzymes, particularly glucosylceramide synthase (GCS), are upregulated during sorafenib treatment in hepatoma cells (HepG2 and Hep3B), and more importantly, in sorafenib-resistant cell lines. GCS silencing or pharmacological GCS inhibition sensitized hepatoma cells to sorafenib exposure. GCS inhibition, combined with sorafenib, triggered cytochrome c release and ATP depletion in sorafenib-treated hepatoma cells, leading to mitochondrial cell death after energetic collapse. Conversely, genetic GCS overexpression increased sorafenib resistance. Of interest, GCS inhibition improved sorafenib effectiveness in a xenograft mouse model, recovering drug sensitivity of sorafenib-resistant tumors in mice. In conclusion, our results reveal GCS induction as a mechanism of sorafenib resistance, suggesting that GCS targeting may be a novel strategy to increase sorafenib efficacy in HCC management, and point to target the mitochondria as the subcellular location where sorafenib therapy could be potentiated.


SDS-PAGE and immunoblot analysis
Cell lysates were prepared in RIPA buffer plus proteinase inhibitors. Samples containing 10 to 30 μg were separated by 8-10% SDS-PAGE. Proteins were transferred to nitrocellulose membranes, blocked in 8% nonfat milk for 1h at room temperature, and incubated overnight at 4°C with the primary antibodies: BECLIN-1 (SantaCruz, sc-11427)

MTT assay
Cytotoxicity of sorafenib and other reagents were determined by the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5diphenyl tetrazolium bromide) assay. About 3×10 4 cells per well were plated into a 96-well plate and incubated in 5% CO2 at 37°C. Stock solution (5 mg/ml) of sterile filtered MTT in phosphate buffered saline (PBS) pH 7.4 was added 1/10 to each well. After 2 h of incubation at 37°C, unreacted dye was removed by aspiration, the insoluble formazan crystals were dissolved in 100 μl/well of solubilization/stop solution (1-propanol) and measured spectrophotometrically in an ELISA reader (Titertek Plus MS 212, ICN, Eschwege, Germany) at a wavelength of 570 nm, and 630 nm (reference). The spectrophotometer was calibrated to zero absorbance using 1-propanol. The net A570 -A630 was taken as the index of cell viability. The net absorbance change taken from the wells of untreated cultured cells was used as the 100% viability value. The relative cell viability (%) related to control wells was calculated by the formula 100X [(A570-A630) sample/(A570-A630)control].

Crystal violet assay
About 5x10 4 cells per well were plated into a 12well plate and incubated in 5% CO2 at 37°C. After four days of treatments, cells were fixed for 5 min. with 10% formalin, stained for 30 minutes and washed twice with tap water. After draining, plates were inverted for 5 minutes and photographed.

Caspase-3 activity determination
Cells were treated with different extracellular factors, media was removed and cells were scrapped in a buffer containing 120 mM NaCl, 50 mM Tris-HCl, pH7.4, 0.5 % Igepal, 2 mM ethylene glycol-bis-(2aminoethyl ether) tetraacetic acid, and 50 μM PMSF, incubated at 4°C for 15 minutes with shaking, and spun down at 12,000 g at 4°C for 15 minutes. Caspase activity was assayed with 200 μg of cell lysate and 1.25 ml of assay buffer containing 100 mM NaCl, 10% sucrose, 0.1% (3-[(3-cholamidopropyl)dimethylammonio]-1propanesulfonate (CHAPS), and 10 mM DTT, pH 7.4, by the release of 7-aminomethyl coumarin (AMC) from 40 nmol of Ac-DEVD-AMC (Calbiochem). Fluorescence was continuously recorded with emission at 460 nm and excitation at 380 nm. A unit of caspase-3 activity is the amount of active enzyme necessary to produce an increase in 1 fluorescence unit in spectrofluorimeter. Results are usually represented as Arbitrary Unit/h/μg protein. Data are expressed as fold induction versus control.

Analysis of mitochondrial transmembrane potencial by JC-1 staining assay
Cells were loaded with C5,5ʹ,6,6ʹ-tetrachloro-1,1ʹ,3,3ʹ-tetraethylbenzimidazolyl-carbocyanine iodide (JC-1), a cationic dye that exhibits potential-dependent accumulation in mitochondria, indicated by a fluorescence emission shift from green (~525 nm) to red (~590 nm). Consequently, mitochondrial depolarization is indicated by a decrease in the red/green fluorescence intensity ratio. To evaluate the mitochondrial depolarization induced by drug treatment, we plated 5×10 5 Hep3B cells in 6 well plates. The following day cells were treated as indicated in each experiment. After the treatment, cells were stained for 20 minutes in medium containing JC-1 at the final concentration of 5 μg/ml. After removal of JC-1, cells were washed once by PBS. Fluorescence is measured in a Microplate Fluorescence Reader using an exciting wavelength of 550 (535) nm for red and 485 nm for green, and an emission wavelength of 600 (590) nm and 535 (530) nm, respectively. Results are calculated as a ratio of red fluorescence divided by green fluorescence. The ratio of red to green fluorescence is decreased in dead cells and in cells undergoing apoptosis compared to healthy cells.

Hoechst staining
About 5×10 4 cells per well were plated into a 12well plate and incubated in 5% CO 2 at 37°C. After six hours of treatments, 1/1000 of a Hoechst 33258 (10 mg/ ml stock solution) was added and incubated for 30 min. Cells were washed twice with regular medium. Images were taken under UV light and 12 random fields for each condition were counted to establish the percentage of cells with condensed nucleus.

Reactive Oxygen Species (ROS) measurement
In order to measure the intracellular content of ROS, we used fluorimetric method. Cells were plated in 12-well plates, and after the treatment they were washed with PBS and incubated with 10 μM 2ʹ, 7'-dichlorodihydrofluorescein diacetate (H2DCFDA) in HBSS during 30 minutes at 37°C. The color was incorporated into the cells and converted into 2ʹ,7ʹdichlorofluorescein (DCF) when oxidized by hydrogen peroxide. Fluorescence is measured in a Microplate Fluorescence Reader using an exciting wavelength of 485 nm and an emission wavelength of 520 nm. Results are calculated as fluorescence units per μg protein and expressed as percentage of control.

ATP determination
To determine the intracellular ATP content, we used a bioluminescent assay from Sigma-Aldrich. In short, hepatoma cells were plated in 96-well plates. After the treatments, culture medium was removed, 90 μL of ATP reagent added to each well and the plate tapped briefly to mix. After 1 minute incubation at room temperature, light emission was determined using a luminometer and images of the plate taken with ImageQuant LAS 4000 equipment (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). Results are calculated as light units per μg protein and expressed as percentage of control.

GAPDH activity assay
To determine the intracellular GAPDH activity we used a spectrophotometric kit from Biovision (#K680-100). In short, through this assay GAPDH catalyzes conversion of GAP into 1, 3-Bisphosphate Glycerate and an intermediate, which reacts with the developer to form a colored product (maximal Abs. 450 nm). Results, after standard comparison, are expressed as GAPDH μUnits per ml and minute.

Complex I activity
For the analysis of mitochondrial OXPHOS Complex I enzyme activity from hepatoma protein extracts we used the Complex I Enzyme Activity Dipstick Assay Kit (ab109720) following the manufacturer's instructions. In brief, 20 μg of protein extract from treated cells were loaded onto a dipstick for each sample. Individual dipsticks were incubated with activity buffer solution containing NADH as a substrate and nitrotetrazolium blue (NBT) as the electron acceptor, and developed for 45 minutes. After that, bands were photographed and quantified.

[ 14 C] labeled ceramide quantification by TLC
Hepatocellular carcinoma cells Hep3B were cultured in Dulbecco's modified Eagle medium supplemented with 10% heat inactivated fetal bovine serum. Cells were cultured and labeled with [ 14 C] palmitic acid (2 μCi/ml) (Amersham, Piscataway, NJ) for 24h. After the treatment with Sorafenib/PDMP/NOE, cells were washed three times with phosphate buffered saline, scraped off the dish and centrifuged at 6000 rpm for 5 min. Total lipids were extracted from the pellets with 1 ml chloroform/ methanol (2:1 by volume), 4h. Ceramide was resolved in LK6D gel 60A TLC plates using chloroform/acetic acid (90:10). Ceramides were separated from the other lipids by thin layer chromatography using diethylether/methanol (99:2, v:v) as the developing system. Two separate spots of radiolabeled lipids were detected on the chromatogram by radioactive scanning.

GCS enzyme assay
The hepatocarcinoma cells were seeded in 6 well plates at 5×10 5 /well. After treatment the samples were scraped off the dish with TE plus 0.1% TritonX-100 pH=7.4, vortexed, centrifuged at 12000 rpm for 5 minutes. Supernatant protein (150 mg/ml) was added to a mix of C6-NBD (1μL/sample), UDP-Glucose (5μL/sample) and assay buffer Phosphate 0.1% Triton pH=7.8, for the final volume of 250uL per each sample. Samples were incubated for 1h at 37°C, added 750uL of chloroform:methanol (2:1, v:v), vortexed and centrifuged at 5000 rpm for 5 min. The top (aqueous) layer of each tube is then removed by aspiration and discarded. The bottom layer was transferred to new tubes and dried with heat at the speed-vac. Samples were then redissolved in 50uL of spotting solvent, methanol:chloroform (1:1, v:v), and run on Silica 60 thin-layer chromatography plates in chloroform:methanol:ammonia (90:20:0.5 mL) along with appropriate C6-NBD-lipid standards. The fluorescent lipids on the plates were visualized by UV illumination and quantified using an Alphaimager 2200 Analysis System (Alexandria, VA) and results confirmed in a Perkin Elmer LS50B luminescence spectrometer set at 475 nm for excitation and 525 nm for emission.

ACDase enzyme assay
The cells were seeded in 6 well plates at 5x10 5 cell/well concentration. After treatments, the samples were lysed in an acidic buffer (pH 4.5) consisting of 50 mM sodium acetate, 5 mM magnesium chloride, 1mM EDTA, and 0.5% TritonX-100, and samples centrifuged at 12000rpm for 5 minutes. Supernatant protein (150 μg) was incubated for 1h at 37°C in the assay buffer (pH 4.5) with 0.2% Igepal-CA 630, 250 mM sodium acetate, and 150 μM of C12-NBD (1μL/sample), for the final volume of 250 μL per each sample. Then, 750 μL of chloroform:methanol (2:1, v:v) was added to each tube, vortexed, and centrifuged at 5000 rpm for 5 min. The top (aqueous) layer of each tube was removed by aspiration and discarded. The bottom layer was transferred to new tubes and dried with heat at the speed-vac. Samples were then redissolved in 50uL of spotting solvent, chloroform:methanol (1:1, v:v), and run on Silica 60 thin-layer chromatography plates in chloroform:methanol:ammonia (90:20:0.5) along with appropriate C12-NBD-lipid standards. The fluorescent lipids on the plates were visualized by UV illumination and quantified using an Alphaimager 2200 Analysis System (Alexandria, VA) and results confirmed in a Perkin Elmer LS50B luminescence spectrometer set at 475 nm for excitation and 525 nm for emission.

cDNA array
TissueScan™ cDNA Array (Liver Cancer cDNA Array I, Origene) was used to quantify GCS levels in tumor and normal tissues. Tissue cDNAs of each array are synthesized from high quality total RNAs of pathologistverified tissues, normalized and validated with ß-actin in two sequential qPCR analyses, and provided with clinical information and QC data. Our array contained cDNA from 48 samples covering 8-normal, 7-Stage I, 8-II, 8-IIIA, 3-IV and 13-Liver Lesions in identical plates (LVRT101).