Growth arrest DNA damage-inducible gene 45 gamma expression as a prognostic and predictive biomarker in hepatocellular carcinoma.

Growth arrest DNA damage-inducible gene 45 (GADD45) family proteins play a crucial role in regulating cellular stress responses and apoptosis. The present study explored the prognostic and predictive role of GADD45γ in hepatocellular carcinoma (HCC) treatment. GADD45γ expression in HCC cells was examined using quantitative reverse transcription-PCR (qRT-PCR) and Western blotting. The control of GADD45γ transcription was examined using a luciferase reporter assay and chromatin immunoprecipitation. The in vivo induction of GADD45γ was performed using adenoviral transfer. The expression of GADD45γ in HCC tumor tissues from patients who had undergone curative resection was measured using qRT-PCR. Sorafenib induced expression of GADD45γ mRNA and protein, independent of its RAF kinase inhibitor activity. GADD45γ induction was more prominent in sorafenib-sensitive HCC cells (Huh-7 and HepG2, IC50 6-7 μM) than in sorafenib-resistant HCC cells (Hep3B, Huh-7R, and HepG2R, IC50 12-15 μM). Overexpression of GADD45γ reversed sorafenib resistance in vitro and in vivo, whereas GADD45γ expression knockdown by using siRNA partially abrogated the proapoptotic effects of sorafenib on sorafenib-sensitive cells. Overexpression of survivin in HCC cells abolished the antitumor enhancement between GADD45γ overexpression and sorafenib treatment, suggesting that survivin is a crucial mediator of antitumor effects of GADD45γ. GADD45γ expression decreased in tumors from patients with HCC who had undergone curative surgery, and low GADD45γ expression was an independent prognostic factor for poor survival, in addition to old age and vascular invasion. The preceding data indicate that GADD45γ suppression is a poor prognostic factor in patients with HCC and may help predict sorafenib efficacy in HCC.


Reporter constructs and assay of the GADD45γ promoter activity
Proximal promoter fragments of GADD45γ, spanning -1711 to +14, were cloned upstream of the luciferase gene in the pGL4.17-base luciferase expression plasmid (Promega). Six different GADD45γ promoter deletion fragments were generated by PCR using the sense and antisense primers listed in supplement. An EcoRV site was incorporated into the sense primers, and a HindIII site was incorporated into the antisense primers. The TFSEARCH program (version 1.3; http:// www.cbrc.jp.reserach/db/TFSEARCH.html) was used to identify possible binding sites for transcription factors in the GADD45γ promoter. Gene synthesis was performed to obtain sequences that mutated at CEBP (CEBPm) binding sites (sequences listed in supplementary table 1) from Genomics BioScience and Technology Co., Ltd. (Taiwan). Huh-7 genomic DNA was used as the PCR template. PCR products were digested, purified, and cloned into the corresponding sites of pGL4.17 vector (Promega, Madison, WI). Huh-7 cells were transfected with individual GADD45γ reporter constructs using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) and co-transfected with pGL4.73 [hRluc/SV40], which constitutively expresses renilla luciferase, to normalize transfection efficiency. The promoter activities with or without sorafenib treatment were determined by dual-luciferase assay kit (Promega), according to the manufacturer's directions. The activity levels were expressed relative to a vector control.

Cell viability and apoptosis assays
Cell viability was assessed using an MTT (3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide) assay and the IC 50 values of individual drugs were calculated using CompuSyn software (ComboSyn, Paramus, NJ) based on the changes of absorbance measured by spectrophometry (DTX 880; Beckman Coulter, Fullerton, CA) as previously described [1]. The fraction of apoptotic cells after drug treatment was assessed by sub-G1 fraction analysis and Annexin V staining using flow cytometry as previously described [1].

Annexin V assay
The fraction of apoptotic cells after drug treatment was assessed by Annexin V analysis using flow cytometry. HCC cells were treated with the drugs for 48 h at indicated concentrations, and the cells were trypsinized and collected. For Annexin V analysis, the cells were centrifuged at 300 × g for 5 min at room temperature, and the cell suspension was stained with Annexin V-FITC (Annexin V assay kit, BD Biosciences Pharmingen) and propidium iodide at room temperature for at least 15 min in the dark. The cells were then analyzed by FACScan flow cytometer and Cell Quest program. The proportion of apoptotic cells was the proportion of cells stained with Annexin V. Columns, mean of three independent experiments; bars, SD. *, p < 0.05; **, p < 0.01.

Western blot analysis
Whole cell lysates of HCC cells after drug treatment were prepared and quantified. Nuclear and cytoplasmic fractions were extracted with a CMN Compartment Protein Extraction kit (BioChain, Hayward, CA). SDS-PAGE and Western blot analysis were performed to measure protein expression. Signals were visualized using a UVP Imaging System (UVP, Upland, CA) or with X-ray film. A recombinant adenovirus (Ad-GADD45γ) was generated by homologous recombination and amplified in human embryonic kidney 293 cells. The adenovirus with no insert (Ad-empty) was used as a control. Viruses were purified by CsCl density gradient centrifugation, and viral titer was determined as previously described [2,3]. Purified virus was stored in 10 mmol/l Tris-HCl (pH 8.0), 2 mmol/l MgCl2, and 4% (vol/vol) sucrose at -80°C until used for the experiments. For infection, Huh-7R cells (~3 × 10 6 ) were infected 24 hours before inoculation into mice with the Ad-GADD45γ or Ad-empty virus at the 10 multiplicity of infection (10 MOI).  Representative survival curves were shown below.

Overexpression of GADD45γ and survivin in vitro and in vivo
The case set consisted of 190 HCC samples that have mRNA, CAN, and sequencing data (accessed on 25 January 2015). A sensitivity test using Z score from 0.3 to 2.0 was done. In Z scores 0.3 to 0.8, both up-and down-regulation were classified as 'alterations' in GADD45γ expression, as shown below, and it was thus difficult to compare survival in patients with or without alterations in GADD45γ expression according to the instructions from the cBio Portal website.