Epigenetic footprint enables molecular risk stratification of hepatoblastoma with clinical implications.

BACKGROUND & AIMS: Hepatoblastoma (HB) is a rare disease. Nevertheless, it is the predominant pediatric liver cancer, with limited therapeutic options for patients with aggressive tumors. Herein, we aimed to uncover the mechanisms of HB pathobiology and to identify new biomarkers and therapeutic targets in a move towards precision medicine for patients with advanced HB. METHODS: We performed a comprehensive genomic, transcriptomic and epigenomic characterization of 159 clinically annotated samples from 113 patients with HB, using high-throughput technologies. RESULTS: We discovered a widespread epigenetic footprint of HB that includes hyperediting of the tumor suppressor BLCAP concomitant with a genome-wide dysregulation of RNA editing and the overexpression of mainly non-coding genes of the oncogenic 14q32 DLK1-DIO3 locus. By unsupervised analysis, we identified 2 epigenomic clusters (Epi-CA, Epi-CB) with distinct degrees of DNA hypomethylation and CpG island hypermethylation that are associated with the C1/C2/C2B transcriptomic subtypes. Based on these findings, we defined the first molecular risk stratification of HB (MRS-HB), which encompasses 3 main prognostic categories and improves the current clinical risk stratification approach. The MRS-3 category (28%), defined by strong 14q32 locus expression and Epi-CB methylation features, was characterized by CTNNB1 and NFE2L2 mutations, a progenitor-like phenotype and clinical aggressiveness. Finally, we identified choline kinase alpha as a promising therapeutic target for intermediate and high-risk HBs, as its inhibition in HB cell lines and patient-derived xenografts strongly abrogated tumor growth. CONCLUSIONS: These findings provide a detailed insight into the molecular features of HB and could be used to improve current clinical stratification approaches and to develop treatments for patients with HB. LAY SUMMARY: Hepatoblastoma is a rare childhood liver cancer that has been understudied. We have used cutting-edge technologies to expand our molecular knowledge of this cancer. Our biological findings can be used to improve clinical management and pave the way for the development of novel therapies for this cancer.

CD5L is upregulated in hepatocellular carcinoma and promotes liver cancer cell proliferation and antiapoptotic responses by binding to HSPA5 (GRP78).

CD5-like (CD5L) is a soluble scavenger cysteine-rich protein that modulates inflammatory responses. We studied the involvement of CD5L in liver cancer. Immunohistochemistry (IHC) of CD5L in 60 hepatocellular carcinomas and 34 adjacent nontumor livers, showed that CD5L staining was higher in tumor than in nontumor tissue (Mann-Whitney test; P = 0.0039). High CD5L correlated with elevated proliferation (Ki67, linear regression; P < 0.0001) and lower patient event-free survival (log-rank; P = 0.0185). Accordingly, CD5L expression was detected in the liver cancer cell lines Huh7, HepG2, and SNU-398. In vitro technologies using these cell lines, including small interfering RNA (siRNA) and cDNA transfection, showed that CD5L promoted colony formation and cell proliferation and protected against cisplatin-induced apoptosis. To find a molecular explanation for these roles, novel CD5L-interacting protein ligands in liver cancer cells were identified by immunoprecipitation followed by mass spectrometry. Among these, the molecular chaperone of the unfolded protein response (UPR), heat shock protein (HSP)-A5, was selected for validation. The interaction was confirmed by confocal microscopy in the Huh7 and HepG2 cell lines. Furthermore, functional experiments revealed that CD5L activates the UPR and autophagy mechanisms in Huh7 cells, thereby providing a novel molecular link between the UPR and autophagy in liver cancer.-Aran, G., Sanjurjo, L., Bárcena, C., Simon-Coma, M., Téllez, É., Vázquez-Vitali, M., Garrido, M., Guerra, L., Díaz, E., Ojanguren, I., Elortza, F., Planas, R., Sala, M., Armengol, C., Sarrias, M.-R. CD5L is upregulated in hepatocellular carcinoma and promotes liver cancer cell proliferation and antiapoptotic responses by binding to HSPA5 (GRP78).

Optimizing anaerobic growth rate and fermentation kinetics in Saccharomyces cerevisiae strains expressing Calvin-cycle enzymes for improved ethanol yield.

BACKGROUND: Reduction or elimination of by-product formation is of immediate economic relevance in fermentation processes for industrial bioethanol production with the yeast Saccharomyces cerevisiae. Anaerobic cultures of wild-type S. cerevisiae require formation of glycerol to maintain the intracellular NADH/NAD+ balance. Previously, functional expression of the Calvin-cycle enzymes ribulose-1,5-bisphosphate carboxylase (RuBisCO) and phosphoribulokinase (PRK) in S. cerevisiae was shown to enable reoxidation of NADH with CO2 as electron acceptor. In slow-growing cultures, this engineering strategy strongly decreased the glycerol yield, while increasing the ethanol yield on sugar. The present study explores engineering strategies to improve rates of growth and alcoholic fermentation in yeast strains that functionally express RuBisCO and PRK, while maximizing the positive impact on the ethanol yield. RESULTS: Multi-copy integration of a bacterial-RuBisCO expression cassette was combined with expression of the Escherichia coli GroEL/GroES chaperones and expression of PRK from the anaerobically inducible DAN1 promoter. In anaerobic, glucose-grown bioreactor batch cultures, the resulting S. cerevisiae strain showed a 31% lower glycerol yield and a 31% lower specific growth rate than a non-engineered reference strain. Growth of the engineered strain in anaerobic, glucose-limited chemostat cultures revealed a negative correlation between its specific growth rate and the contribution of the Calvin-cycle enzymes to redox homeostasis. Additional deletion of GPD2, which encodes an isoenzyme of NAD+-dependent glycerol-3-phosphate dehydrogenase, combined with overexpression of the structural genes for enzymes of the non-oxidative pentose-phosphate pathway, yielded a CO2-reducing strain that grew at the same rate as a non-engineered reference strain in anaerobic bioreactor batch cultures, while exhibiting a 86% lower glycerol yield and a 15% higher ethanol yield. CONCLUSIONS: The metabolic engineering strategy presented here enables an almost complete elimination of glycerol production in anaerobic, glucose-grown batch cultures of S. cerevisiae, with an associated increase in ethanol yield, while retaining near wild-type growth rates and a capacity for glycerol formation under osmotic stress. Using current genome-editing techniques, the required genetic modifications can be introduced in one or a few transformations. Evaluation of this concept in industrial strains and conditions is therefore a realistic next step towards its implementation for improving the efficiency of first- and second-generation bioethanol production.