Heterochromatin formation and remodeling by IRTKS condensates counteract cellular senescence.

Heterochromatin, a key component of the eukaryotic nucleus, is fundamental to the regulation of genome stability, gene expression and cellular functions. However, the factors and mechanisms involved in heterochromatin formation and maintenance still remain largely unknown. Here, we show that insulin receptor tyrosine kinase substrate (IRTKS), an I-BAR domain protein, is indispensable for constitutive heterochromatin formation via liquid‒liquid phase separation (LLPS). In particular, IRTKS droplets can infiltrate heterochromatin condensates composed of HP1α and diverse DNA-bound nucleosomes. IRTKS can stabilize HP1α by recruiting the E2 ligase Ubc9 to SUMOylate HP1α, which enables it to form larger phase-separated droplets than unmodified HP1α. Furthermore, IRTKS deficiency leads to loss of heterochromatin, resulting in genome-wide changes in chromatin accessibility and aberrant transcription of repetitive DNA elements. This leads to activation of cGAS-STING pathway and type-I interferon (IFN-I) signaling, as well as to the induction of cellular senescence and senescence-associated secretory phenotype (SASP) responses. Collectively, our findings establish a mechanism by which IRTKS condensates consolidate constitutive heterochromatin, revealing an unexpected role of IRTKS as an epigenetic mediator of cellular senescence.

BAIAP2L1 and BAIAP2L2 differently regulate hair cell stereocilia morphology.

Inner ear sensory hair cells are characterized by their apical F-actin-based cell protrusions named stereocilia. In each hair cell, several rows of stereocilia with different height are organized into a staircase-like pattern. The height of stereocilia is tightly regulated by two protein complexes, namely row-1 and row-2 tip complex, that localize at the tips of tallest-row and shorter-row stereocilia, respectively. Previously, we and others identified BAI1-associated protein 2-like 2 (BAIAP2L2) as a component of row-2 complex that play an important role in maintaining shorter-row stereocilia. In the present work we show that BAIAP2L1, an ortholog of BAIAP2L2, localizes at the tips of tallest-row stereocilia in a way dependent on known row-1 complex proteins EPS8 and MYO15A. Interestingly, unlike BAIAP2L2 whose stereocilia-tip localization requires calcium, the localization of BAIAP2L1 on the tips of tallest-row stereocilia is calcium-independent. Therefore, our data suggest that BAIAP2L1 and BAIAP2L2 localize at the tips of different stereociliary rows and might regulate the development and/or maintenance of stereocilia differently. However, loss of BAIAP2L1 does not affect the row-1 protein complex, and the auditory and balance function of Baiap2l1 knockout mice are largely normal. We hypothesize that other orthologous protein(s) such as BAIAP2 might compensate for the loss of BAIAP2L1 in the hair cells.

LZP is required for hepatic triacylglycerol transportation through maintaining apolipoprotein B stability.

The conserved zona pellucida (ZP) domain is found in hundreds of extracellular proteins that are expressed in various organs and play a variety of roles as structural components, receptors and tumor suppressors. A liver-specific zona pellucida domain-containing protein (LZP), also named OIT3, has been shown to be mainly expressed in human and mouse hepatocytes; however, the physiological function of LZP in the liver remains unclear. Here, we show that Lzp deletion inhibited very low-density lipoprotein (VLDL) secretion, leading to hepatic TG accumulation and lower serum TG levels in mice. The apolipoprotein B (apoB) levels were significantly decreased in the liver, serum, and VLDL particles of LZP-deficient mice. In the presence of LZP, which is localized to the endoplasmic reticulum (ER) and Golgi apparatus, the ER-associated degradation (ERAD) of apoB was attenuated; in contrast, in the absence of LZP, apoB was ubiquitinated by AMFR, a known E3 ubiquitin ligase specific for apoB, and was subsequently degraded, leading to lower hepatic apoB levels and inhibited VLDL secretion. Interestingly, hepatic LZP levels were elevated in mice challenged with a high-fat diet and humans with simple hepatic steatosis, suggesting that LZP contributes to the physiological regulation of hepatic TG homeostasis. In general, our data establish an essential role for LZP in hepatic TG transportation and VLDL secretion by preventing the AMFR-mediated ubiquitination and degradation of apoB and therefore provide insight into the molecular function of LZP in hepatic lipid metabolism.

Investigating higher-order interactions in single-cell data with scHOT.

Single-cell genomics has transformed our ability to examine cell fate choice. Examining cells along a computationally ordered 'pseudotime' offers the potential to unpick subtle changes in variability and covariation among key genes. We describe an approach, scHOT-single-cell higher-order testing-which provides a flexible and statistically robust framework for identifying changes in higher-order interactions among genes. scHOT can be applied for cells along a continuous trajectory or across space and accommodates various higher-order measurements including variability or correlation. We demonstrate the use of scHOT by studying coordinated changes in higher-order interactions during embryonic development of the mouse liver. Additionally, scHOT identifies subtle changes in gene-gene correlations across space using spatially resolved transcriptomics data from the mouse olfactory bulb. scHOT meaningfully adds to first-order differential expression testing and provides a framework for interrogating higher-order interactions using single-cell data.

DLK1-directed chimeric antigen receptor T-cell therapy for hepatocellular carcinoma.

BACKGROUND: Delta-like homologue 1 (DLK1), a transmembrane protein, is highly expressed in hepatocellular carcinoma (HCC). We explored whether DLK1-directed chimeric antigen receptor (CAR) T cells can specifically eliminate DLK1-positive HCC cells and serve as a therapeutic strategy for HCC immunotherapy. METHODS: We first characterized a homemade anti-human DLK1 monoclonal antibody, sequenced the single-chain Fragment variable (scFv) and integrated it into the second-generation CAR lentiviral vector, and then developed the DLK1-directed CAR-T cells. The cytotoxic activities of DLK1-directed CAR-T cells against different HCC cells were evaluated in vitro and in vivo. RESULTS: The genetically modified human T cells with the DLK1-directed CARs produced cytotoxic activity against DLK1-positive HCC cells. Additionally, the DLK1-directed CARs enhanced T cell proliferation and activation in a DLK1-dependent manner. Interestingly, the DLK1-targeted CAR-T cells significantly inhibited both subcutaneous and peritoneal xenograft tumours derived from human liver cancer cell lines HepG2 or Huh-7. CONCLUSION: DLK1-directed CAR-T cells specifically suppresses DLK1-positive HCC cells in vitro and in vivo. This study provides a novel transmembrane antigen DLK1 as a potential therapeutic target appropriate for CAR-T cell therapy, which may be further developed as a clinical therapeutic strategy for HCC immunotherapy.

scMerge leverages factor analysis, stable expression, and pseudoreplication to merge multiple single-cell RNA-seq datasets.

Concerted examination of multiple collections of single-cell RNA sequencing (RNA-seq) data promises further biological insights that cannot be uncovered with individual datasets. Here we present scMerge, an algorithm that integrates multiple single-cell RNA-seq datasets using factor analysis of stably expressed genes and pseudoreplicates across datasets. Using a large collection of public datasets, we benchmark scMerge against published methods and demonstrate that it consistently provides improved cell type separation by removing unwanted factors; scMerge can also enhance biological discovery through robust data integration, which we show through the inference of development trajectory in a liver dataset collection.

DeepAntigen: a novel method for neoantigen prioritization via 3D genome and deep sparse learning.

MOTIVATION: The mutations of cancers can encode the seeds of their own destruction, in the form of T-cell recognizable immunogenic peptides, also known as neoantigens. It is computationally challenging, however, to accurately prioritize the potential neoantigen candidates according to their ability of activating the T-cell immunoresponse, especially when the somatic mutations are abundant. Although a few neoantigen prioritization methods have been proposed to address this issue, advanced machine learning model that is specifically designed to tackle this problem is still lacking. Moreover, none of the existing methods considers the original DNA loci of the neoantigens in the perspective of 3D genome which may provide key information for inferring neoantigens' immunogenicity. RESULTS: In this study, we discovered that DNA loci of the immunopositive and immunonegative MHC-I neoantigens have distinct spatial distribution patterns across the genome. We therefore used the 3D genome information along with an ensemble pMHC-I coding strategy, and developed a group feature selection-based deep sparse neural network model (DNN-GFS) that is optimized for neoantigen prioritization. DNN-GFS demonstrated increased neoantigen prioritization power comparing to existing sequence-based approaches. We also developed a webserver named deepAntigen (http://yishi.sjtu.edu.cn/deepAntigen) that implements the DNN-GFS as well as other machine learning methods. We believe that this work provides a new perspective toward more accurate neoantigen prediction which eventually contribute to personalized cancer immunotherapy. AVAILABILITY AND IMPLEMENTATION: Data and implementation are available on webserver: http://yishi.sjtu.edu.cn/deepAntigen. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The Mutational Features of Aristolochic Acid-Induced Mouse and Human Liver Cancers.

BACKGROUND AND AIMS: Aristolochic acid (AA) exposure has been statistically associated with human liver cancers. However, direct evidence of AA exposure-induced liver cancer is absent. This study aims to establish a direct causal relationship between AA exposure and liver cancers based on a mouse model and then explores the AA-mediated genomic alterations that could be implicated in human cancers with AA-associated mutational signature. APPROACH AND RESULTS: We subjected mice, including phosphatase and tensin homolog (Pten)-deficient ones, to aristolochic acid I (AAI) alone or a combination of AAI and CCl4 . Significantly, AAI exposure induced mouse liver cancers, including hepatocellular carcinoma (HCC) and combined HCC and intrahepatic cholangiocarcinoma, in a dose-dependent manner. Moreover, AAI exposure also enhanced tumorigenesis in these CCl4 -treated or Pten-deficient mice. AAI led to DNA damage and AAI-DNA adduct that could initiate liver cancers through characteristic adenine-to-thymine transversions, as indicated by comprehensive genomic analysis, which revealed recurrent mutations in Harvey rat sarcoma virus oncogene. Interestingly, an AA-associated mutational signature was mainly implicated in human liver cancers, especially from China. Moreover, we detected the AAI-DNA adduct in 25.8% (16/62) of paratumor liver tissues from randomly selected Chinese patients with HCC. Furthermore, based on phylogenetic analysis, the characteristic mutations were found in the initiating malignant clones in the AA-implicated mouse and human liver cancers where the mutations of tumor protein p53 and Janus kinase 1 were prone to be significantly enriched in the AA-affected human tumors. CONCLUSIONS: This study provides evidence for AA-induced liver cancer with the featured mutational processes during malignant clonal evolution, laying a solid foundation for the prevention and diagnosis of AA-associated human cancers, especially liver cancers.

Mutational and transcriptomic landscapes of a rare human prostate basal cell carcinoma.

BACKGROUND: As a rare subtype of prostate carcinoma, basal cell carcinoma (BCC) has not been studied extensively and thus lacks systematic molecular characterization. METHODS: Here, we applied single-cell genomic amplification and RNA-Seq to a specimen of human prostate BCC (CK34ßE12+ /P63+ /PAP- /PSA- ). The mutational landscape was obtained via whole exome sequencing of the amplification mixture of 49 single cells, and the transcriptomes of 69 single cells were also obtained. RESULTS: The five putative driver genes mutated in BCC are CASC5, NUTM1, PTPRC, KMT2C, and TBX3, and the top three nucleotide substitutions are C>T, T>C, and C>A, similar to common prostate cancer. The distribution of the variant allele frequency values indicated that these single cells are from the same tumor clone. The 69 single cells were clustered into tumor, stromal, and immune cells based on their global transcriptomic profiles. The tumor cells specifically express basal cell markers like KRT5, KRT14, and KRT23 and epithelial markers EPCAM, CDH1, and CD24. The transcription factor covariance network analysis showed that the BCC tumor cells have distinct regulatory networks. By comparison with current prostate cancer datasets, we found that some of the bulk samples exhibit basal cell signatures. Interestingly, at single-cell resolution the gene expression patterns of prostate BCC tumor cells show uniqueness compared with that of common prostate cancer-derived circulating tumor cells. CONCLUSIONS: This study, for the first time, discloses the comprehensive mutational and transcriptomic landscapes of prostate BCC, which lays a foundation for the understanding of its tumorigenesis mechanism and provides new insights into prostate cancers in general.

Optimal Gene Filtering for Single-Cell data (OGFSC)-a gene filtering algorithm for single-cell RNA-seq data.

MOTIVATION: Single-cell transcriptomic data are commonly accompanied by extremely high technical noise due to the low RNA concentrations from individual cells. Precise identification of differentially expressed genes and cell populations are heavily dependent on the effective reduction of technical noise, e.g. by gene filtering. However, there is still no well-established standard in the current approaches of gene filtering. Investigators usually filter out genes based on single fixed threshold, which commonly leads to both over- and under-stringent errors. RESULTS: In this study, we propose a novel algorithm, termed as Optimal Gene Filtering for Single-Cell data, to construct a thresholding curve based on gene expression levels and the corresponding variances. We validated our method on multiple single-cell RNA-seq datasets, including simulated and published experimental datasets. The results show that the known signal and known noise are reliably discriminated in the simulated datasets. In addition, the results of seven experimental datasets demonstrate that these cells of the same annotated types are more sharply clustered using our method. Interestingly, when we re-analyze the dataset from an aging research recently published in Science, we find a list of regulated genes which is different from that reported in the original study, because of using different filtering methods. However, the knowledge based on our findings better matches the progression of immunosenescence. In summary, we here provide an alternative opportunity to probe into the true level of technical noise in single-cell transcriptomic data. AVAILABILITY AND IMPLEMENTATION: https://github.com/XZouProjects/OGFSC.git. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Conditional knockin of Dnmt3a R878H initiates acute myeloid leukemia with mTOR pathway involvement.

DNMT3A is frequently mutated in acute myeloid leukemia (AML). To explore the features of human AML with the hotspot DNMT3A R882H mutation, we generated Dnmt3a R878H conditional knockin mice, which developed AML with enlarged Lin-Sca1+cKit+ cell compartments. The transcriptome and DNA methylation profiling of bulk leukemic cells and the single-cell RNA sequencing of leukemic stem/progenitor cells revealed significant changes in gene expression and epigenetic regulatory patterns that cause differentiation arrest and growth advantage. Consistent with leukemic cell accumulation in G2/M phase, CDK1 was up-regulated due to mTOR activation associated with DNA hypomethylation. Overexpressed CDK1-mediated EZH2 phosphorylation resulted in an abnormal trimethylation of H3K27 profile. The mTOR inhibitor rapamycin elicited a significant therapeutic response in Dnmt3aR878H/WT mice.

[A method for reliable detection of genomic point mutations based on single-cell target-sequencing].

The analysis of genomic point mutations is one of the research strategies to explore the clonal evolution of tumor cells. At present, clonal evolution of tumor cells is mainly determined by bulk sampling and sequencing of different sections of the tumor. Since this approach analyzes a mixture of different cell types, it may not accurately represent the clonal evolution of specific tumor cell populations and likely miss low frequency mutations, especially when the sequencing depths are not sufficient. To address this issue, we have developed a strategy to analyze genomic point mutations from prostate basal cell carcinoma (BCC) tissues at single-cell resolution. Firstly, we optimized the single-cell whole genome amplification procedure with HepG2 cells. Then the single cells from BCC tissue were captured by a microfluidic chip of Fluidigm and processed for whole-genome amplification. Both SCUBE3 and MST1L genomic mutations were obtained by whole exome sequencing. Finally, we examined the genomic mutations through single-cell targeted amplification and Sanger sequencing. The established method successfully reconfirmed the mutations of SCUBE3 and MST1L in BCC at single cell level. The strategy established in this study could provide a useful tool for determining the clonal evolution of tumor cells based on genomic mutations at single-cell resolution.

Application of qPCR assays based on haloacids transporter gene dehp2 for discrimination of Burkholderia and Paraburkholderia.

BACKGROUND: A major facilitator superfamily transporter Dehp2 was recently shown to be playing an important role in transport and biodegradation of haloacids in Paraburkholderia caribensis MBA4, and Dehp2 is phylogenetically conserved in Burkholderia sensu lato. RESULTS: We designed both Burkholderia sensu stricto-specific and Paraburkholderia-specific qPCR assays based on dehp2 and 16S rRNA, and validated the qPCR assays in 12 bacterial strains. The qPCR assays could detect single species of Burkholderia sensu stricto or Paraburkholderia with high sensitivity and discriminate them in mixtures with high specificity over a wide dynamic range of relative concentrations. At relatively lower cost compared with sequencing-based approach, the qPCR assays will facilitate discrimination of Burkholderia sensu stricto and Paraburkholderia in a large number of samples. CONCLUSIONS: For the first time, we report the utilization of a haloacids transporter gene for discriminative purpose in Burkholderia sensu lato. This enables not only quick decision on proper handling of putative pathogenic samples in Burkholderia sensu stricto group but also future exploitation of relevant species in Paraburkholderia group for haloacids biodegradation purposes.

IRTKS is correlated with progression and survival time of patients with gastric cancer.

BACKGROUND AND OBJECTIVES: IRTKS functions as a novel regulator of tumour suppressor p53; however, the role of IRTKS in pathogenesis of gastric cancer is unclear. DESIGN: We used immunohistochemistry to detect IRTKS levels in 527 human gastric cancer specimens. We generated both IRTKS-deficient and p53-deficient mice to observe survival time of these mice and to isolate mouse embryonic fibroblasts (MEFs) for evaluating in vivo tumorigenicity. Co-immunoprecipitation was used to study the interaction among p53, MDM2 and IRTKS, as well as the ubiquitination of p53. RESULTS: IRTKS was significantly overexpressed in human gastric cancer, which was conversely associated with wild-type p53 expression. Among patients with wild-type p53 (n=206), those with high IRTKS expression (n=141) had a shorter survival time than those with low IRTKS (n=65) (p=0.0153). Heterozygous p53+/- mice with IRTKS deficiency exhibited significantly delayed tumorigenesis and an extended tumour-free survival time. p53+/- MEFs without IRTKS exhibited attenuated in vivo tumorigenicity. IRTKS depletion upregulated p53 and its target genes, such as BAX and p21. Intriguingly, IRTKS overexpression promoted p53 ubiquitination and degradation in MEFs and gastric cancer cells. Under DNA damage conditions, IRTKS was phosphorylated at Ser331 by the activated Chk2 kinase and then dissociated from p53, along with the p53-specific E3 ubiquitin ligase MDM2, resulting in attenuated p53 ubiquitination and degradation. CONCLUSION: IRTKS overexpression is negatively correlated with progression and overall survival time of patients with gastric cancer with wild-type p53 through promotion of p53 degradation via the ubiquitin/proteasome pathway.

Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks.

BACKGROUND: With the developments of DNA sequencing technology, large amounts of sequencing data have been produced that provides unprecedented opportunities for advanced association studies between somatic mutations and cancer types/subtypes which further contributes to more accurate somatic mutation based cancer typing (SMCT). In existing SMCT methods however, the absence of high-level feature extraction is a major obstacle in improving the classification performance. RESULTS: We propose DeepCNA, an advanced convolutional neural network (CNN) based classifier, which utilizes copy number aberrations (CNAs) and HiC data, to address this issue. DeepCNA first pre-process the CNA data by clipping, zero padding and reshaping. Then, the processed data is fed into a CNN classifier, which extracts high-level features for accurate classification. Experimental results on the COSMIC CNA dataset indicate that 2D CNN with both cell lines of HiC data lead to the best performance. We further compare DeepCNA with three widely adopted classifiers, and demonstrate that DeepCNA has at least 78% improvement of performance. CONCLUSIONS: This paper demonstrates the advantages and potential of the proposed DeepCNA model for processing of somatic point mutation based gene data, and proposes that its usage may be extended to other complex genotype-phenotype association studies.

Single-cell RNA-Seq analysis reveals dynamic trajectories during mouse liver development.

BACKGROUND: The differentiation and maturation trajectories of fetal liver stem/progenitor cells (LSPCs) are not fully understood at single-cell resolution, and a priori knowledge of limited biomarkers could restrict trajectory tracking. RESULTS: We employed marker-free single-cell RNA-Seq to characterize comprehensive transcriptional profiles of 507 cells randomly selected from seven stages between embryonic day 11.5 and postnatal day 2.5 during mouse liver development, and also 52 Epcam-positive cholangiocytes from postnatal day 3.25 mouse livers. LSPCs in developing mouse livers were identified via marker-free transcriptomic profiling. Single-cell resolution dynamic developmental trajectories of LSPCs exhibited contiguous but discrete genetic control through transcription factors and signaling pathways. The gene expression profiles of cholangiocytes were more close to that of embryonic day 11.5 rather than other later staged LSPCs, cuing the fate decision stage of LSPCs. Our marker-free approach also allows systematic assessment and prediction of isolation biomarkers for LSPCs. CONCLUSIONS: Our data provide not only a valuable resource but also novel insights into the fate decision and transcriptional control of self-renewal, differentiation and maturation of LSPCs.

ARID1A represses hepatocellular carcinoma cell proliferation and migration through lncRNA MVIH.

ARID1A, encoding the BAF250a subunit of SWI/SNF complex, has a high mutation frequency in numerous types of cancer. LncRNAs, a type of non-coding RNAs longer than 200 nucleotides, have been reported to interplay with SWI/SNF complex during cancer progression. However, whether the interaction between ARID1A and lncRNA affects hepatocellular carcinoma (HCC) still needs to be investigated. Here, we reveal that ARID1A interacts with lncRNA MVIH through some region(s) or domain(s) including ARID domain and C-terminal ARID1A protein binding domain. ARID1A upregulates its downstream target CDKN1A and suppresses HCC cell proliferation and migration through inhibiting MVIH. Our data suggests that deficiency or loss of functional mutations of ARID1A in HCC cells might contribute to the increased activity of certain cancer-promoting lncRNAs.

NOXIN as a cofactor of DNA polymerase-primase complex could promote hepatocellular carcinoma.

Oncogene activation or inactivation of tumor suppressor genes are crucial to tumor initiation and progression. DNA copy number amplification is one of many mechanisms that activate oncogenes in many tumors, including hepatocellular carcinoma (HCC). Although it has been known that some oncogenes such as c-myc amplification is involved in HCC pathogenesis, more oncogenes with DNA copy amplification contribute to HCC initiation and progression remain to be characterized. Here, we identified NOXIN as a novel potential oncogene with DNA copy number amplification by Single Nucleotide Polymorphism microarray-based genome-wide DNA copy number analysis of 43 human HCC samples. We identified the focal DNA gain and amplification region containing NOXIN on chromosome 11q14.1 and NOXIN overexpression significantly associated with HCC progression. We then assessed the role of NOXIN in HCC cells. NOXIN overexpression promoted cellular proliferation, colony formation, cellular migration and in vivo tumorigenicity, whereas NOXIN knockdown attenuated these effects. Interestingly, NOXIN overexpression accelerated the G1-S phase transition by enhancing DNA synthesis. Furthermore, we found that NOXIN interacts with DNA polymerase α, suggesting that NOXIN may promote de novo DNA synthesis by promoting DNA polymerase-primase complex formation. These collective data indicated that NOXIN overexpression, as a result of genomic DNA gain or amplification, promotes HCC tumorigenesis by accelerating DNA synthesis and cell cycle progression, where NOXIN functions as a cofactor of DNA polymerase-primase complex by associating with DNA polymerase α.

Functional genomic studies: insights into the pathogenesis of liver cancer.

Liver cancer is the sixth-most-common cancer overall but the third-most-frequent cause of cancer death. Among primary liver cancers, hepatocellular carcinoma (HCC), the major histological subtype, is associated with multiple risk factors, including hepatitis B and C virus infection, alcohol consumption, obesity, and diet contamination. Although previous studies have revealed that certain genetic and epigenetic changes, such as TP53 and ß-catenin mutations, occur in HCC cells, the pathogenesis of this cancer remains obscure. Functional genomic approaches-including genome-wide association studies, whole-genome and whole-exome sequencing, array-based comparative genomic hybridization, global DNA methylome mapping, and gene or noncoding RNA expression profiling-have recently been applied to HCC patients with different clinical features to uncover the genetic risk factors and underlying molecular mechanisms involved in this cancer's initiation and progression. The genome-wide analysis of germline and somatic genetic and epigenetic events facilitates understanding of the pathogenesis and molecular classification of liver cancer as well as the identification of novel diagnostic biomarkers and therapeutic targets for cancer.