Additional Supporting Information may be found in the online version of this article.


Supporting Fig.1. Relative induction of cytochromes P450s (CYPs) in the MFH model. Total RNAs were extracted from MFH cultures at the indicated times, and the mRNA expression of Cypla2, Cyp3all, Cyp3a41, and Cyp3a44 was determined by RT-qPCR. The housekeeping gene Gapdh was used as an internal control to normalize the amount of cDNA.

Supporting Fig. 2. Fetal and postnatal expression of the Dnmt and miR-148a/b/152 family in the mouse liver. (A) Sequence homology of the three conserved mature miRNAs. The blue shading highlights the conserved bases among species and the different miRNAs. The red letters specify the seed sequences. (B) Time course of miR-148a/b/152 expression during liver development. (C) Time course of Dnmtl, Dnmt3a, and Dnmt3b mRNA level during embryonic development. Fetal and adult livers were collected and physically dissociated at the indicated embryonic stages. Before total RNA extraction, tissue homogenates were treated with a red blood cell lysis buffer. MiRNA and mRNA expression was assessed by RT-qPCR and normalized against the endogenous controls RNU6B and Gapdh.

Supporting Fig. 3. Methylation status of the miR-148a and albumin promoters. (A) Combined bisulfite restriction analysis (COBRA) and genomic bisulfite sequencing of the miR-148a promoter region during the differentiation of MFHs. U, unmethylated; M, methylated. Each row represents a single clone for each individual genome. Open and filled squares represent unmethylated and methylated CpG sites, respectively. Arrows show CpG sites on which COBRA was performed. (B) Methylation status of the albumin promoter determined by COBRA in the MFH model. (C) COBRA assay of the albumin gene promoter in HepG2 cells treated with or without 5-Aza for 5 days and normal human hepatocytes. (D) Epigenetic modification of the albumin promoter region 48 and 72 h after transfection of HepG2 cells with miR-148a mimics and controls.

Supporting Fig. 4. Effect of miR-148a overexpression on CYP mRNA levels in MFHs. Primary cultures of MFHs were transfected using miR-148a mimics or controls at 4 days after their purification (MFH D4). Total RNAs were extracted at the indicated times following transfection; then, Cyp3a1 1, Cyp3a41, and Cyp3a44 expression was determined by RT-qPCR. Statistical significance was reached at **P < 0.01 and ***P < 0.001 (t-test).

Supporting Fig. 5.Expression of miR-148a regarding the hepatitis virus infection type in HCC patients. (A) RT-qPCR analysis showing the differential expression of miR-148a in 13 normal livers (NL) and 39 primary HCC tumors (HCC) and their corresponding non-tumor tissues (NT). All HCC were related to HBV (n = 18) or HCV infection (n = 21). The expression of miR-148a was normalized with RNU6B. The boxes show the median and 25th and 75th percentiles, whereas vertical bars display the range of values. Within the box, 50% of values are shown, and 80% are included between the extremities of vertical bars. The Mann-Whitney U-test indicated statistical under-expression of miR-148a in HCCs and adjacent NT tissues in both HBV and HCV cohorts as compared with the normal liver group (P < 0.001). (B) Paired representation of miR148a expression between primary tumors and their adjacent non-tumor areas in HBV-HCC and HCV-HCC clinical samples. Statistical differences were analyzed with the Wilcoxon signed-rank test and indicated significant inhibition for the tumor versus the NT group in HBV-HCC pairs (P = 0.0268). MiR-148a expression was not significantly inhibited in the HBV group.

Supporting Fig. 6. Evaluation of circulating miR-148a levels in the serum of patients with HCC recurrence. Samples were collected in two steps from 11 HCC patients exhibiting HCV infection: i) after surgical resection of the primary tumor and ii) following the diagnosis of HCC recurrence. Synthetic C. ele2ans miR-39 was used as an invariant control for 200 µL of serum subjected to phenol extraction. (A) Paired representation of serum miR-148a levels between blood samples collected following tumor resection (post-surgery) and HCC recurrence. (B) Serum miR-148a expression analyses of the 11 HCC patients (ratio recurrence/post-surgery). A diminution of circulating miR-148a was observed in 8 patients following recurrence (P = 0.2783, Wilcoxon signed-rank test).

Supporting Fig.7. Evaluation of miR-148a overexpression effect on apoptosis activity in the mouse Hepa 1-6 and human HepG2 cell lines. (A) Hepa 1-6 caspase 3/7 activity assessed at the indicated times following cell transfection with miR-148a mimics and negative controls. As Hepa 1-6 cells did not respond to TNF-related apoptosis-inducing„ li2and (TRAIL) treatment (data not shown), (B) caspase activity assay was performed using the human cell line HepG2. The day following transfection with miR-148a mimics or controls, cells were treated with the apoptosis inducer TRAIL at 50 ng/mL or left untreated. Then, Caspase 3/7 activities were assessed 5 h and 10 h after the beginning of treatment. No statistical difference was found for all experiments (t- test).

Supporting Fig. 8. (A) Relative expression of the oncoaene c-Met in the Hepa 1-6 cell line, undifferentiated MFHs, and mature-induced hepatocytes. Messenger RNA levels of c-Met were found to be significantly increased in mouse HCC cells and undifferentiated MFHs (MFH D2). As compared with induced-mature hepatocytes (MFH D8), Hepa 1-6 exhibited a 2.77-fold overexpression of c-Met (P < 0.01, t-test). (B) 3'-UTR assay in the Hepa 1-6 cell line. The sequences indicate the interaction sites between miR-148a and the 3'-UTR of c-Met in mouse and human species. Dual luciferase activities were measured after Hepa 1-6 co-transfection with miR-148a mimics and the firefly-Renilla luciferase construct containing the mouse c-Met 3'- UTR sequence. Cells were collected 24 h after transfection, and the firetly/Renilla luciferase activities were determined. The empty vector was used as a negative control. No difference in the luciferase activities was statistically found. A similar result was observed using a human c-MET 3'-UTR construct (data not shown).


Supporting Table 1. Microarray data. (see attached Excel file)

Supporting Table 2. Clinical data from HCC patients, normal livers, and fetal livers.

Supporting Table 3. List of primers used for RT-qPCR.

Supporting Table 4. List of primers used for COBRA.

hep26422-sup-0003-suppinfo.doc125KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.