Pharmacological Targeting of STK19 Inhibits Oncogenic NRAS-Driven Melanomagenesis.

Activating mutations in NRAS account for 20%-30% of melanoma, but despite decades of research and in contrast to BRAF, no effective anti-NRAS therapies have been forthcoming. Here, we identify a previously uncharacterized serine/threonine kinase STK19 as a novel NRAS activator. STK19 phosphorylates NRAS to enhance its binding to its downstream effectors and promotes oncogenic NRAS-mediated melanocyte malignant transformation. A recurrent D89N substitution in STK19 whose alterations were identified in 25% of human melanomas represents a gain-of-function mutation that interacts better with NRAS to enhance melanocyte transformation. STK19D89N knockin leads to skin hyperpigmentation and promotes NRASQ61R-driven melanomagenesis in vivo. Finally, we developed ZT-12-037-01 (1a) as a specific STK19-targeted inhibitor and showed that it effectively blocks oncogenic NRAS-driven melanocyte malignant transformation and melanoma growth in vitro and in vivo. Together, our findings provide a new and viable therapeutic strategy for melanomas harboring NRAS mutations.

Cytoplasmic SIRT6-mediated ACSL5 deacetylation impedes nonalcoholic fatty liver disease by facilitating hepatic fatty acid oxidation.

Nonalcoholic fatty liver disease (NAFLD) is characterized by excessive hepatic lipid accumulation, which can progress to nonalcoholic steatohepatitis (NASH). Histone deacetylase Sirtuin 6 (SIRT6) regulates NAFLD by regulating metabolism-related gene expression, but an extrachromosomal role for SIRT6 in NAFLD development remains elusive. We investigated whether SIRT6 functions on NAFLD in the cytoplasm. We found that SIRT6 binds saturated fatty acids, especially palmitic acid. This binding leads to its nuclear export, where it deacetylates long-chain acyl-CoA synthase 5 (ACSL5), thereby facilitating fatty acid oxidation. High-fat diet-induced NAFLD is suppressed by ACSL5 hepatic overexpression but is exacerbated by its depletion. As confirmation, overexpression of a deacetylated ACSL5 mimic attenuated NAFLD in Sirt6 liver-specific knockout mice. Moreover, NASH-hepatic tissues from both patients and diet-fed mice exhibited significantly reduced cytoplasmic SIRT6 levels and increased ACSL5 acetylation. The SIRT6/ACSL5 signaling pathway has a critical role in NAFLD progression and might constitute an avenue for therapeutic intervention.

The GATAD2B-NuRD complex drives DNA:RNA hybrid-dependent chromatin boundary formation upon DNA damage.

Double-strand breaks (DSBs) are the most lethal form of DNA damage. Transcriptional activity at DSBs, as well as transcriptional repression around DSBs, are both required for efficient DNA repair. The chromatin landscape defines and coordinates these two opposing events. However, how the open and condensed chromatin architecture is regulated remains unclear. Here, we show that the GATAD2B-NuRD complex associates with DSBs in a transcription- and DNA:RNA hybrid-dependent manner, to promote histone deacetylation and chromatin condensation. This activity establishes a spatio-temporal boundary between open and closed chromatin, which is necessary for the correct termination of DNA end resection. The lack of the GATAD2B-NuRD complex leads to chromatin hyperrelaxation and extended DNA end resection, resulting in homologous recombination (HR) repair failure. Our results suggest that the GATAD2B-NuRD complex is a key coordinator of the dynamic interplay between transcription and the chromatin landscape, underscoring its biological significance in the RNA-dependent DNA damage response.

C1QBP Promotes Homologous Recombination by Stabilizing MRE11 and Controlling the Assembly and Activation of MRE11/RAD50/NBS1 Complex.

MRE11 nuclease forms a trimeric complex (MRN) with RAD50 and NBS1 and plays a central role in preventing genomic instability. When DNA double-strand breaks (DSBs) occur, MRN is quickly recruited to the damage site and initiates DNA end resection; accordingly, MRE11 must be tightly regulated to avoid inefficient repair or nonspecific resection. Here, we show that MRE11 and RAD50 form a complex (MRC) with C1QBP, which stabilizes MRE11/RAD50, while inhibiting MRE11 nuclease activity by preventing its binding to DNA or chromatin. Upon DNA damage, ATM phosphorylates MRE11-S676/S678 to quickly dissociate the MRC complex. Either excess or insufficient C1QBP impedes the recruitment of MRE11 to DSBs and impairs the DNA damage response. C1QBP is highly expressed in breast cancer and positively correlates with MRE11 expression, and the inhibition of C1QBP enhances tumor regression with chemotherapy. By influencing MRE11 at multiple levels, C1QBP is, thus, an important player in the DNA damage response.

Acetylation of PHF5A Modulates Stress Responses and Colorectal Carcinogenesis through Alternative Splicing-Mediated Upregulation of KDM3A.

Alternative pre-mRNA-splicing-induced post-transcriptional gene expression regulation is one of the pathways for tumors maintaining proliferation rates accompanying the malignant phenotype under stress. Here, we uncover a list of hyperacetylated proteins in the context of acutely reduced Acetyl-CoA levels under nutrient starvation. PHF5A, a component of U2 snRNPs, can be acetylated at lysine 29 in response to multiple cellular stresses, which is dependent on p300. PHF5A acetylation strengthens the interaction among U2 snRNPs and affects global pre-mRNA splicing pattern and extensive gene expression. PHF5A hyperacetylation-induced alternative splicing stabilizes KDM3A mRNA and promotes its protein expression. Pathologically, PHF5A K29 hyperacetylation and KDM3A upregulation axis are correlated with poor prognosis of colon cancer. Our findings uncover a mechanism of an anti-stress pathway through which acetylation on PHF5A promotes the cancer cells' capacity for stress resistance and consequently contributes to colon carcinogenesis.

Multiscale topology in interactomic network: from transcriptome to antiaddiction drug repurposing.

The escalating drug addiction crisis in the United States underscores the urgent need for innovative therapeutic strategies. This study embarked on an innovative and rigorous strategy to unearth potential drug repurposing candidates for opioid and cocaine addiction treatment, bridging the gap between transcriptomic data analysis and drug discovery. We initiated our approach by conducting differential gene expression analysis on addiction-related transcriptomic data to identify key genes. We propose a novel topological differentiation to identify key genes from a protein-protein interaction network derived from DEGs. This method utilizes persistent Laplacians to accurately single out pivotal nodes within the network, conducting this analysis in a multiscale manner to ensure high reliability. Through rigorous literature validation, pathway analysis and data-availability scrutiny, we identified three pivotal molecular targets, mTOR, mGluR5 and NMDAR, for drug repurposing from DrugBank. We crafted machine learning models employing two natural language processing (NLP)-based embeddings and a traditional 2D fingerprint, which demonstrated robust predictive ability in gauging binding affinities of DrugBank compounds to selected targets. Furthermore, we elucidated the interactions of promising drugs with the targets and evaluated their drug-likeness. This study delineates a multi-faceted and comprehensive analytical framework, amalgamating bioinformatics, topological data analysis and machine learning, for drug repurposing in addiction treatment, setting the stage for subsequent experimental validation. The versatility of the methods we developed allows for applications across a range of diseases and transcriptomic datasets.

Lost in translation: How neurons cope with tRNA decoding.

Post-transcriptional tRNA modifications contribute to the decoding efficiency of tRNAs by supporting codon recognition and tRNA stability. Recent work shows that the molecular and cellular functions of tRNA modifications and tRNA-modifying-enzymes are linked to brain development and neurological disorders. Lack of these modifications affects codon recognition and decoding rate, promoting protein aggregation and translational stress response pathways with toxic consequences to the cell. In this review, we discuss the peculiarity of local translation in neurons, suggesting a role for fine-tuning of translation performed by tRNA modifications. We provide several examples of tRNA modifications involved in physiology and pathology of the nervous system, highlighting their effects on protein translation and discussing underlying mechanisms, like the unfolded protein response (UPR), ribosome quality control (RQC), and no-go mRNA decay (NGD), which could affect neuronal functions. We aim to deepen the understanding of the roles of tRNA modifications and the coordination of these modifications with the protein translation machinery in the nervous system.

Genetically Encoded Lizard Color Divergence for Camouflage and Thermoregulation.

Local adaptation is critical in speciation and evolution, yet comprehensive studies on proximate and ultimate causes of local adaptation are generally scarce. Here, we integrated field ecological experiments, genome sequencing, and genetic verification to demonstrate both driving forces and molecular mechanisms governing local adaptation of body coloration in a lizard from the Qinghai-Tibet Plateau. We found dark lizards from the cold meadow population had lower spectrum reflectance but higher melanin contents than light counterparts from the warm dune population. Additionally, the colorations of both dark and light lizards facilitated the camouflage and thermoregulation in their respective microhabitat simultaneously. More importantly, by genome resequencing analysis, we detected a novel mutation in Tyrp1 that underpinned this color adaptation. The allele frequencies at the site of SNP 459# in the gene of Tyrp1 are 22.22% G/C and 77.78% C/C in dark lizards and 100% G/G in light lizards. Model-predicted structure and catalytic activity showed that this mutation increased structure flexibility and catalytic activity in enzyme TYRP1, and thereby facilitated the generation of eumelanin in dark lizards. The function of the mutation in Tyrp1 was further verified by more melanin contents and darker coloration detected in the zebrafish injected with the genotype of Tyrp1 from dark lizards. Therefore, our study demonstrates that a novel mutation of a major melanin-generating gene underpins skin color variation co-selected by camouflage and thermoregulation in a lizard. The resulting strong selection may reinforce adaptive genetic divergence and enable the persistence of adjacent populations with distinct body coloration.

Artificial intelligence-aided protein engineering: from topological data analysis to deep protein language models.

Protein engineering is an emerging field in biotechnology that has the potential to revolutionize various areas, such as antibody design, drug discovery, food security, ecology, and more. However, the mutational space involved is too vast to be handled through experimental means alone. Leveraging accumulative protein databases, machine learning (ML) models, particularly those based on natural language processing (NLP), have considerably expedited protein engineering. Moreover, advances in topological data analysis (TDA) and artificial intelligence-based protein structure prediction, such as AlphaFold2, have made more powerful structure-based ML-assisted protein engineering strategies possible. This review aims to offer a comprehensive, systematic, and indispensable set of methodological components, including TDA and NLP, for protein engineering and to facilitate their future development.

TripletCell: a deep metric learning framework for accurate annotation of cell types at the single-cell level.

Single-cell RNA sequencing (scRNA-seq) has significantly accelerated the experimental characterization of distinct cell lineages and types in complex tissues and organisms. Cell-type annotation is of great importance in most of the scRNA-seq analysis pipelines. However, manual cell-type annotation heavily relies on the quality of scRNA-seq data and marker genes, and therefore can be laborious and time-consuming. Furthermore, the heterogeneity of scRNA-seq datasets poses another challenge for accurate cell-type annotation, such as the batch effect induced by different scRNA-seq protocols and samples. To overcome these limitations, here we propose a novel pipeline, termed TripletCell, for cross-species, cross-protocol and cross-sample cell-type annotation. We developed a cell embedding and dimension-reduction module for the feature extraction (FE) in TripletCell, namely TripletCell-FE, to leverage the deep metric learning-based algorithm for the relationships between the reference gene expression matrix and the query cells. Our experimental studies on 21 datasets (covering nine scRNA-seq protocols, two species and three tissues) demonstrate that TripletCell outperformed state-of-the-art approaches for cell-type annotation. More importantly, regardless of protocols or species, TripletCell can deliver outstanding and robust performance in annotating different types of cells. TripletCell is freely available at https://github.com/liuyan3056/TripletCell. We believe that TripletCell is a reliable computational tool for accurately annotating various cell types using scRNA-seq data and will be instrumental in assisting the generation of novel biological hypotheses in cell biology.

Machine Learning Methods for Small Data Challenges in Molecular Science.

Small data are often used in scientific and engineering research due to the presence of various constraints, such as time, cost, ethics, privacy, security, and technical limitations in data acquisition. However, big data have been the focus for the past decade, small data and their challenges have received little attention, even though they are technically more severe in machine learning (ML) and deep learning (DL) studies. Overall, the small data challenge is often compounded by issues, such as data diversity, imputation, noise, imbalance, and high-dimensionality. Fortunately, the current big data era is characterized by technological breakthroughs in ML, DL, and artificial intelligence (AI), which enable data-driven scientific discovery, and many advanced ML and DL technologies developed for big data have inadvertently provided solutions for small data problems. As a result, significant progress has been made in ML and DL for small data challenges in the past decade. In this review, we summarize and analyze several emerging potential solutions to small data challenges in molecular science, including chemical and biological sciences. We review both basic machine learning algorithms, such as linear regression, logistic regression (LR), k-nearest neighbor (KNN), support vector machine (SVM), kernel learning (KL), random forest (RF), and gradient boosting trees (GBT), and more advanced techniques, including artificial neural network (ANN), convolutional neural network (CNN), U-Net, graph neural network (GNN), Generative Adversarial Network (GAN), long short-term memory (LSTM), autoencoder, transformer, transfer learning, active learning, graph-based semi-supervised learning, combining deep learning with traditional machine learning, and physical model-based data augmentation. We also briefly discuss the latest advances in these methods. Finally, we conclude the survey with a discussion of promising trends in small data challenges in molecular science.

The HDAC6-RNF168 axis regulates H2A/H2A.X ubiquitination to enable double-strand break repair.

Histone deacetylase 6 (HDAC6) mediates DNA damage signaling by regulating the mismatch repair and nucleotide excision repair pathways. Whether HDAC6 also mediates DNA double-strand break (DSB) repair is unclear. Here, we report that HDAC6 negatively regulates DSB repair in an enzyme activity-independent manner. In unstressed cells, HDAC6 interacts with H2A/H2A.X to prevent its interaction with the E3 ligase RNF168. Upon sensing DSBs, RNF168 rapidly ubiquitinates HDAC6 at lysine 116, leading to HDAC6 proteasomal degradation and a restored interaction between RNF168 and H2A/H2A.X. H2A/H2A.X is ubiquitinated by RNF168, precipitating the recruitment of DSB repair factors (including 53BP1 and BRCA1) to chromatin and subsequent DNA repair. These findings reveal novel regulatory machinery based on an HDAC6-RNF168 axis that regulates the H2A/H2A.X ubiquitination status. Interfering with this axis might be leveraged to disrupt a key mechanism of cancer cell resistance to genotoxic damage and form a potential therapeutic strategy for cancer.

GATA3 functions downstream of BRCA1 to promote DNA damage repair and suppress dedifferentiation in breast cancer.

BACKGROUND: Inadequate DNA damage repair promotes aberrant differentiation of mammary epithelial cells. Mammary luminal cell fate is mainly determined by a few transcription factors including GATA3. We previously reported that GATA3 functions downstream of BRCA1 to suppress aberrant differentiation in breast cancer. How GATA3 impacts DNA damage repair preventing aberrant cell differentiation in breast cancer remains elusive. We previously demonstrated that loss of p18, a cell cycle inhibitor, in mice induces luminal-type mammary tumors, whereas depletion of either Brca1 or Gata3 in p18 null mice leads to basal-like breast cancers (BLBCs) with activation of epithelial-mesenchymal transition (EMT). We took advantage of these mutant mice to examine the role of Gata3 as well as the interaction of Gata3 and Brca1 in DNA damage repair in mammary tumorigenesis. RESULTS: Depletion of Gata3, like that of Brca1, promoted DNA damage accumulation in breast cancer cells in vitro and in basal-like breast cancers in vivo. Reconstitution of Gata3 improved DNA damage repair in Brca1-deficient mammary tumorigenesis. Overexpression of GATA3 promoted homologous recombination (HR)-mediated DNA damage repair and restored HR efficiency of BRCA1-deficient cells. Depletion of Gata3 sensitized tumor cells to PARP inhibitor (PARPi), and reconstitution of Gata3 enhanced resistance of Brca1-deficient tumor cells to PARP inhibitor. CONCLUSIONS: These results demonstrate that Gata3 functions downstream of BRCA1 to promote DNA damage repair and suppress dedifferentiation in mammary tumorigenesis and progression. Our findings suggest that PARP inhibitors are effective for the treatment of GATA3-deficient BLBCs.

Poisson-Boltzmann-based machine learning model for electrostatic analysis.

Electrostatics is of paramount importance to chemistry, physics, biology, and medicine. The Poisson-Boltzmann (PB) theory is a primary model for electrostatic analysis. However, it is highly challenging to compute accurate PB electrostatic solvation free energies for macromolecules due to the nonlinearity, dielectric jumps, charge singularity, and geometric complexity associated with the PB equation. The present work introduces a PB-based machine learning (PBML) model for biomolecular electrostatic analysis. Trained with the second-order accurate MIBPB solver, the proposed PBML model is found to be more accurate and faster than several eminent PB solvers in electrostatic analysis. The proposed PBML model can provide highly accurate PB electrostatic solvation free energy of new biomolecules or new conformations generated by molecular dynamics with much reduced computational cost.

Small RNAs contribute to citrus Huanglongbing tolerance by manipulating methyl salicylate signaling and exogenous methyl salicylate primes citrus groves from emerging infection.

Citrus production is severely threatened by the devastating Huanglongbing (HLB) disease globally. By studying and analyzing the defensive behaviors of an HLB-tolerant citrus cultivar 'Shatangju', we discovered that citrus can sense Candidatus Liberibacter asiaticus (CLas) infection and induce immune responses against HLB, which can be further strengthened by both endogenously produced and exogenously applied methyl salicylate (MeSA). This immune circuit is turned on by an miR2977-SAMT (encoding a citrus Salicylate [SA] O-methyltransferase) cascade, by which CLas infection leads to more in planta MeSA production and aerial emission. We provided both transgenic and multi-year trail evidences that MeSA is an effective community immune signal. Ambient MeSA accumulation and foliage application can effectively induce defense gene expression and significantly boost citrus performance. We also found that miRNAs are battle fields between citrus and CLas, and about 30% of the differential gene expression upon CLas infection are regulated by miRNAs. Furthermore, CLas hijacks host key processes by manipulating key citrus miRNAs, and citrus employs miRNAs that coordinately regulate defense-related genes. Based on our results, we proposed that miRNAs and associated components are key targets for engineering or breeding resistant citrus varieties. We anticipate that MeSA-based management, either induced expression or external application, would be a promising tool for HLB control.

Prevention and treatment of HPV-related cancer through a mRNA vaccine expressing APC-targeting antigen.

Persistent human papillomavirus (HPV) infection is associated with multiple malignancies. Developing therapeutic vaccines to eliminate HPV-infected and malignant cells holds significant value. In this study, we introduced a lipid nanoparticle encapsulated mRNA vaccine expressing tHA-mE7-mE6. Mutations were introduced into E6 and E7 of HPV to eliminate their tumourigenicity. A truncated influenza haemagglutinin protein (tHA), which binds to the CD209 receptor on the surface of dendritic cells (DCs), was fused with mE7-mE6 in order to allow efficient uptake of antigen by antigen presenting cells. The tHA-mE7-mE6 (mRNA) showed higher therapeutic efficacy than mE7-mE6 (mRNA) in an E6 and E7+ tumour model. The treatment resulted in complete tumour regression and prevented tumour formation. Strong CD8+ T-cell immune response was induced, contributing to preventing and curing of E6 and E7+ tumour. Antigen-specific CD8+ T were found in spleens, peripheral blood and in tumours. In addition, the tumour infiltration of DC and NK cells were increased post therapy. In conclusion, this study described a therapeutic mRNA vaccine inducing strong anti-tumour immunity in peripheral and in tumour microenvironment, holding promising potential to treat HPV-induced cancer and to prevent cancer recurrence.

Chromosome-level genome assembly of a cliff plant Taihangia rupestris var. ciliata provides insights into its adaptation and demographic history.

BACKGROUND: Cliffs are recognized as one of the most challenging environments for plants, characterized by harsh conditions such as drought, infertile soil, and steep terrain. However, they surprisingly host ancient and diverse plant communities and play a crucial role in protecting biodiversity. The Taihang Mountains, which act as a natural boundary in eastern China, support a rich variety of plant species, including many unique to cliff habitats. However, it is little known how cliff plants adapt to harsh habitats and the demographic history in this region. RESULTS: To better understand the demographic history and adaptation of cliff plants in this area, we analyzed the chromosome-level genome of a representative cliff plant, T. rupestris var. ciliata, which has a genome size of 769.5 Mb, with a scaffold N50 of 104.92 Mb. The rapid expansion of transposable elements may have contributed to the increasing genome and its ability to adapt to unique and challenging cliff habitats. Comparative analysis of the genome evolution between Taihangia and non-cliff plants in Rosaceae revealed a significant expansion of gene families associated with oxidative phosphorylation, which is likely a response to the abiotic stresses faced by cliff plants. This expansion may explain the long-term adaptation of Taihangia to harsh cliff environments. The effective population size of the two varieties has continuously decreased due to climatic fluctuations during the Quaternary period. Furthermore, significant differences in gene expression between the two varieties may explain the varied leaf phenotypes and adaptations to harsh conditions in different natural distributions. CONCLUSION: Our study highlights the extraordinary adaptation of T. rupestris var. ciliata, shedding light on the evolution of cliff plants worldwide.

Bridging Eulerian and Lagrangian Poisson-Boltzmann solvers by ESES.

The Poisson-Boltzmann (PB) model is a widely used electrostatic model for biomolecular solvation analysis. Formulated as an elliptic interface problem, the PB model can be numerically solved on either Eulerian meshes using finite difference/finite element methods or Lagrangian meshes using boundary element methods. Molecular surface generators, which produce the discretized dielectric interfaces between solutes and solvents, are critical factors in determining the accuracy and efficiency of the PB solvers. In this work, we investigate the utility of the Eulerian Solvent Excluded Surface (ESES) software for rendering conjugated Eulerian and Lagrangian surface representations, which enables us to numerically validate and compare the quality of Eulerian PB solvers, such as the MIBPB solver, and the Lagrangian PB solvers, such as the TABI-PB solver. Furthermore, with the ESES software and its associated PB solvers, we are able to numerically validate an interesting and useful but often neglected source-target symmetric property associated with the linearized PB model.

Shape Anisotropy-Governed High-Performance Nanomagnetosol for In Vivo Magnetic Particle Imaging of Lungs.

Caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), coronavirus disease 2019 (COVID-19) has shown extensive lung manifestations in vulnerable individuals, putting lung imaging and monitoring at the forefront of early detection and treatment. Magnetic particle imaging (MPI) is an imaging modality, which can bring excellent contrast, sensitivity, and signal-to-noise ratios to lung imaging for the development of new theranostic approaches for respiratory diseases. Advances in MPI tracers would offer additional improvements and increase the potential for clinical translation of MPI. Here, a high-performance nanotracer based on shape anisotropy of magnetic nanoparticles is developed and its use in MPI imaging of the lung is demonstrated. Shape anisotropy proves to be a critical parameter for increasing signal intensity and resolution and exceeding those properties of conventional spherical nanoparticles. The 0D nanoparticles exhibit a 2-fold increase, while the 1D nanorods have a > 5-fold increase in signal intensity when compared to VivoTrax. Newly designed 1D nanorods displayed high signal intensities and excellent resolution in lung images. A spatiotemporal lung imaging study in mice revealed that this tracer offers new opportunities for monitoring disease and guiding intervention.

Does depressurization of the portal vein before liver transplantation affect the recurrence of HCC? A nested case-control study.

BACKGROUND: Portal hypertension (PHT) has been proven to be closely related to the development of hepatocellular carcinoma (HCC). Whether PHT before liver transplantation (LT) will affect the recurrence of HCC is not clear. METHODS: 110 patients with depressurization of the portal vein (DPV) operations (Transjugular Intrahepatic Portosystemic Shunt-TIPS, surgical portosystemic shunt or/and splenectomy) before LT from a HCC LT cohort, matched with 330 preoperative non-DPV patients; this constituted a nested case-control study. Subgroup analysis was based on the order of DPV before or after the occurrence of HCC. RESULTS: The incidence of acute kidney injury and intra-abdominal bleeding after LT in the DPV group was significantly higher than that in non-DPV group. The 5-year survival rates in the DPV and non-DPV group were 83.4% and 82.7% respectively (P = 0.930). In subgroup analysis, patients in the DPV prior to HCC subgroup may have a lower recurrence rate (4.7% vs.16.8%, P = 0.045) and a higher tumor free survival rate (88.9% vs.74.4%, P = 0.044) after LT under the up-to-date TNMI-II stage, while in TNM III stage, there was no difference for DPV prior to HCC subgroup compared with the DPV after HCC subgroup or the non-DPV group. CONCLUSION: Compared with DPV after HCC, DPV treatment before HCC can reduce the recurrence rate of HCC after early transplantation (TNM I-II). DPV before LT can reduce the recurrence of early HCC.