(A) H&E-stained sections of human HCC with the indicated histopathological variegation (left panels) shown adjacent to corresponding histopathologies in murine liver cancers arising from p53^−/−;myc hepatoblasts (center panels). Anti-CK8 immunohistochemistry of the murine tumors is shown at right.
(B) H&E staining of in situ and subcutaneously retransplanted HCCs derived from p53^−/− ;myc hepatoblasts.

(A) Genome-wide profile of a human HCC harboring an amplification on chromosome 7 containing the c-met gene and three regions amplified on chromosome 11.
(B) Single-probe resolution of chromosome 11, with genes contained within each amplicon depicted below.
(C) Genome-wide profile of an ovarian carcinoma containing the 11q22 amplification.
(D) Single-probe resolution of the 11q22 amplicon, with genes contained within the amplified region depicted below.

(A) Apoptosis measurements (Cell Death Detection ELISA^PLUS kit; Roche) from p53^−/− hepatoblasts double infected with myc + cIAP1 or myc + vector following culture under the indicated serum conditions for 48 hr (left panel). Cells grown to confluence were cultured for another 48 hr, and cell death was measured (right panel). Error bars represent mean ± SD (n = 3) per data point.
(B) Tumor volume measurements at various times following subcutaneous injection of p53^−/− hepatoblasts double infected with myc + cIAP1 or myc + vector into the rear flanks of nude mice (n = 6 for each group). Shown is a representative of three independent experiments, each showing a statistical difference between cIAP1 and control (p < 0.05). Error bars represent ±SD.
(C) Immunoblotting of tumors overexpressing myc-tagged cIAP1 (lanes 8–13) or control vector tumors (lanes 5–7) using antibodies against cIAP1. Samples from cultured myc-tagged cIAP1-expressing hepatoblasts (M, lane 2) or vector alone (V, lane 1) and from 9qA1 amplicon-containing cells (A+, lane 4) were analyzed for comparison. Note that myc-tagged cIAP1 migrates at 75 kDa and endogenous cIAP1 at 65 kDa. A– is lysate from amplicon-negative cells of the same genotype (p53^−/−;myc). Tubulin was used as a loading control.
(D and E) p53^−/− hepatoblasts coexpressing Ras (D) or Akt (E) with cIAP1 or a control vector were monitored for tumorigenicity following subcutaneous injection into nude mice (n = 6 per group). Error bars represent ±SD.
(F) Immunoblotting of lysates derived from hepatoma cells outgrown from a 9qA1-positive p53^−/−;myc tumor transduced with shRNAs targeting cIAP1 and cIAP2 (sh 1+2) or control vectors (V) using antibodies against cIAP1.
(G and H) 9qA1-positive (G) and -negative (H) hepatoma cells expressing cIAP1/2 shRNAs or a control vector were monitored for tumor growth following subcutaneous injection into nude mice. Error bars represent ±SD.

(A) p53^−/−;myc liver progenitor cells were infected with control vector, cIAP1, or Yap or coinfected with Yap + cIAP1 and then transplanted subcutaneously into nude mice (n = 6 per group). Error bars represent ±SD.
(B) GFP imaging of the tumors described in (A).

(A) Schematic diagram showing the generation of in situ liver carcinomas following retroviral transduction of purified E-cadherin⁺ hepato-blasts (see Figure S1).
(B) Analysis of liver sections from mice seeded with GFP-expressing hepatoblasts 1 week postreconstitution. Left, H&E; middle, anti-GFP immunofluorescence; right, DAPI.
(C) External GFP tumor imaging (top panels) or direct imaging of the respective explanted tumor-bearing livers (bottom panels) of mice reconstituted with p53^−/− hepatoblasts transduced with the indicated oncogene.
(D) Survival curves of mice after intrahepatic seeding of p53^−/− liver progenitor cells transduced with the indicated oncogene or control vector.
(E) Explanted murine liver carcinomas (p53^−/−; myc) were grown briefly in culture and then directly injected into the left liver lobe. Shown is a GFP-expressing (left) in situ tumor (right) 42 days postinjection.

(A) ROMA profile of a tumor derived from p53^−/−;Ras embryonic hepatoblasts. Data plotted are the normalized log ratio for each probe (*EST).
(B) Single-probe resolution of chromosome 15 corresponding to the tumor described in (A).
(C) Genome-wide profiles of three independent HCCs (Tu-7, Tu-9, and Tu-13) derived from p53^−/−;myc embryonic hepatoblasts.
(D) Single-probe resolution of chromosome 9qA1 from the tumors described in (C). The minimal overlap region contains the indicated genes.

(A) cIAP1, cIAP2, Yap, and Porimin mRNA levels in murine HCCs that contain the 9qA1 amplicon as determined by quantitative real-time RT-PCR analysis.
(B) Protein lysates from 9qA1-positive (+) or -negative (−) liver cancers and adult mouse liver were immunoblotted with antibodies against cIAP1, cIAP1/2, Yap, and Porimin. * denotes nonspecific bands. Tubulin was used as a loading control.
(C) Quantitative real-time RT-PCR analysis of cIAP1, cIAP2, Yap, and Porimin expression in human HCCs. * denotes a tumor with the 11q22 amplification. Cutoff for increased expression was >2-fold of the expression level in nonneoplastic liver (normal liver).
(D) Immunohistochemistry of an 11q22-positive (top) and -negative (bottom) human HCC using antibodies against the indicated protein.

(A) The proliferation rates of p53^−/−;myc hepatoblasts expressing Yap or a control vector were assessed by the fraction of nuclei incorporating BrdU after a 1 hr pulse.
(B) Protein lysates from two 9qA1 amplicon-positive tumors (+), two amplicon-negative tumors (−), and adult normal mouse liver were immunoblotted with antibodies against Yap and cyclin E. Tubulin was used as a loading control. * denotes a nonspecific band in the liver lysate.
(C–E) The tumorigenicity of p53^−/− liver progenitor cells coexpressing the indicated oncogene (upper left) with a control vector or Yap was assessed by caliper measurement following subcutaneous injection into the rear flanks of nude mice (n = 4 per group). Error bars represent ±SD.
(F) Protein lysates from hepatoma cells (p53^−/−;myc) stably overexpressing Yap were infected with control vector (V) or two different short-hairpin RNAs targeting Yap (sh1 and sh2) and were analyzed for Yap and cyclin E levels. Tubulin was used as a loading control.
(G) Tumorigenicity of 9qA1-positive cells infected with retroviral vectors expressing shRNAs targeting Yap (sh1Yap or sh2Yap) or control vector (n = 6 per group). Error bars represent ±SD.