CoRAL accurately resolves extrachromosomal DNA genome structures with long-read sequencing.

Extrachromosomal DNA (ecDNA) is a central mechanism for focal oncogene amplification in cancer, occurring in approximately 15% of early-stage cancers and 30% of late-stage cancers. EcDNAs drive tumor formation, evolution, and drug resistance by dynamically modulating oncogene copy-number and rewiring gene-regulatory networks. Elucidating the genomic architecture of ecDNA amplifications is critical for understanding tumor pathology and developing more effective therapies. Paired-end short-read (Illumina) sequencing and mapping have been utilized to represent ecDNA amplifications using a breakpoint graph, where the inferred architecture of ecDNA is encoded as a cycle in the graph. Traversals of breakpoint graph have been used to successfully predict ecDNA presence in cancer samples. However, short-read technologies are intrinsically limited in the identification of breakpoints, phasing together of complex rearrangements and internal duplications, and deconvolution of cell-to-cell heterogeneity of ecDNA structures. Long-read technologies, such as from Oxford Nanopore Technologies, have the potential to improve inference as the longer reads are better at mapping structural variants and are more likely to span rearranged or duplicated regions. Here, we propose CoRAL (Complete Reconstruction of Amplifications with Long reads), for reconstructing ecDNA architectures using long-read data. CoRAL reconstructs likely cyclic architectures using quadratic programming that simultaneously optimizes parsimony of reconstruction, explained copy number, and consistency of long-read mapping. CoRAL substantially improves reconstructions in extensive simulations and 10 datasets from previously-characterized cell lines as compared to previous short and long-read based tools. As long-read usage becomes wide-spread, we anticipate that CoRAL will be a valuable tool for profiling the landscape and evolution of focal amplifications in tumors.

PABP Cooperates with the CCR4-NOT Complex to Promote mRNA Deadenylation and Block Precocious Decay.

Multiple deadenylases are known in vertebrates, the PAN2-PAN3 (PAN2/3) and CCR4-NOT (CNOT) complexes, and PARN, yet their differential functions remain ambiguous. Moreover, the role of poly(A) binding protein (PABP) is obscure, limiting our understanding of the deadenylation mechanism. Here, we show that CNOT serves as a predominant nonspecific deadenylase for cytoplasmic poly(A)+ RNAs, and PABP promotes deadenylation while preventing premature uridylation and decay. PAN2/3 selectively trims long tails (>∼150 nt) with minimal effect on transcriptome, whereas PARN does not affect mRNA deadenylation. CAF1 and CCR4, catalytic subunits of CNOT, display distinct activities: CAF1 trims naked poly(A) segments and is blocked by PABPC, whereas CCR4 is activated by PABPC to shorten PABPC-protected sequences. Concerted actions of CAF1 and CCR4 delineate the ∼27 nt periodic PABPC footprints along shortening tail. Our study unveils distinct functions of deadenylases and PABPC, re-drawing the view on mRNA deadenylation and regulation.

Regulation of Poly(A) Tail and Translation during the Somatic Cell Cycle.

Poly(A) tails are critical for mRNA stability and translation. However, recent studies have challenged this view, showing that poly(A) tail length and translation efficiency are decoupled in non-embryonic cells. Using TAIL-seq and ribosome profiling, we investigate poly(A) tail dynamics and translational control in the somatic cell cycle. We find dramatic changes in poly(A) tail lengths of cell-cycle regulatory genes like CDK1, TOP2A, and FBXO5, explaining their translational repression in M phase. We also find that poly(A) tail length is coupled to translation when the poly(A) tail is <20 nucleotides. However, as most genes have >20 nucleotide poly(A) tails, their translation is regulated mainly via poly(A) tail length-independent mechanisms during the cell cycle. Specifically, we find that terminal oligopyrimidine (TOP) tract-containing transcripts escape global translational suppression in M phase and are actively translated. Our quantitative and comprehensive data provide a revised view of translational control in the somatic cell cycle.

PKR is activated by cellular dsRNAs during mitosis and acts as a mitotic regulator.

dsRNA-dependent protein kinase R (PKR) is a ubiquitously expressed enzyme well known for its roles in immune response. Upon binding to viral dsRNA, PKR undergoes autophosphorylation, and the phosphorylated PKR (pPKR) regulates translation and multiple signaling pathways in infected cells. Here, we found that PKR is activated in uninfected cells, specifically during mitosis, by binding to dsRNAs formed by inverted Alu repeats (IRAlus). While PKR and IRAlu-containing RNAs are segregated in the cytosol and nucleus of interphase cells, respectively, they interact during mitosis when nuclear structure is disrupted. Once phosphorylated, PKR suppresses global translation by phosphorylating the α subunit of eukaryotic initiation factor 2 (eIF2α). In addition, pPKR acts as an upstream kinase for c-Jun N-terminal kinase and regulates the levels of multiple mitotic factors such as cyclins A and B and Polo-like kinase 1 and phosphorylation of histone H3. Disruption of PKR activation via RNAi or expression of a transdominant-negative mutant leads to misregulation of the mitotic factors, delay in mitotic progression, and defects in cytokinesis. Our study unveils a novel function of PKR and endogenous dsRNAs as signaling molecules during the mitosis of uninfected cells.

CoRAL accurately resolves extrachromosomal DNA genome structures with long-read sequencing.

Extrachromosomal DNA (ecDNA) is a central mechanism for focal oncogene amplification in cancer, occurring in approximately 15% of early stage cancers and 30% of late-stage cancers. EcDNAs drive tumor formation, evolution, and drug resistance by dynamically modulating oncogene copy-number and rewiring gene-regulatory networks. Elucidating the genomic architecture of ecDNA amplifications is critical for understanding tumor pathology and developing more effective therapies. Paired-end short-read (Illumina) sequencing and mapping have been utilized to represent ecDNA amplifications using a breakpoint graph, where the inferred architecture of ecDNA is encoded as a cycle in the graph. Traversals of breakpoint graph have been used to successfully predict ecDNA presence in cancer samples. However, short-read technologies are intrinsically limited in the identification of breakpoints, phasing together of complex rearrangements and internal duplications, and deconvolution of cell-to-cell heterogeneity of ecDNA structures. Long-read technologies, such as from Oxford Nanopore Technologies, have the potential to improve inference as the longer reads are better at mapping structural variants and are more likely to span rearranged or duplicated regions. Here, we propose CoRAL (Complete Reconstruction of Amplifications with Long reads), for reconstructing ecDNA architectures using long-read data. CoRAL reconstructs likely cyclic architectures using quadratic programming that simultaneously optimizes parsimony of reconstruction, explained copy number, and consistency of long-read mapping. CoRAL substantially improves reconstructions in extensive simulations and 9 datasets from previously-characterized cell-lines as compared to previous short-read-based tools. As long-read usage becomes wide-spread, we anticipate that CoRAL will be a valuable tool for profiling the landscape and evolution of focal amplifications in tumors.

Short poly(A) tails are protected from deadenylation by the LARP1-PABP complex.

Deadenylation generally constitutes the first and pivotal step in eukaryotic messenger RNA decay. Despite its importance in posttranscriptional regulations, the kinetics of deadenylation and its regulation remain largely unexplored. Here we identify La ribonucleoprotein 1, translational regulator (LARP1) as a general decelerator of deadenylation, which acts mainly in the 30-60-nucleotide (nt) poly(A) length window. We measured the steady-state and pulse-chased distribution of poly(A)-tail length, and found that deadenylation slows down in the 30-60-nt range. LARP1 associates preferentially with short tails and its depletion results in accelerated deadenylation specifically in the 30-60-nt range. Consistently, LARP1 knockdown leads to a global reduction of messenger RNA abundance. LARP1 interferes with the CCR4-NOT-mediated deadenylation in vitro by forming a ternary complex with poly(A)-binding protein (PABP) and poly(A). Together, our work reveals a dynamic nature of deadenylation kinetics and a role of LARP1 as a poly(A) length-specific barricade that creates a threshold for deadenylation.

Coordinate regulation of the senescent state by selective autophagy.

Cellular senescence is a complex stress response implicated in aging. Autophagy can suppress senescence but is counterintuitively necessary for full senescence. Although its anti-senescence role is well described, to what extent autophagy contributes to senescence establishment and the underlying mechanisms is poorly understood. Here, we show that selective autophagy of multiple regulatory components coordinates the homeostatic state of senescence. We combined a proteomic analysis of autophagy components with protein stability profiling, identifying autophagy substrate proteins involved in several senescence-related processes. Selective autophagy of KEAP1 promoted redox homeostasis during senescence. Furthermore, selective autophagy limited translational machinery components to ameliorate senescence-associated proteotoxic stress. Lastly, selective autophagy of TNIP1 enhanced senescence-associated inflammation. These selective autophagy networks appear to operate in vivo senescence during human osteoarthritis. Our data highlight a caretaker role of autophagy in the stress support network of senescence through regulated protein stability and unravel the intertwined relationship between two important age-related processes.

The RNA-binding protein repertoire of embryonic stem cells.

RNA-binding proteins (RBPs) have essential roles in RNA-mediated gene regulation, and yet annotation of RBPs is limited mainly to those with known RNA-binding domains. To systematically identify the RBPs of embryonic stem cells (ESCs), we here employ interactome capture, which combines UV cross-linking of RBP to RNA in living cells, oligo(dT) capture and MS. From mouse ESCs (mESCs), we have defined 555 proteins constituting the mESC mRNA interactome, including 283 proteins not previously annotated as RBPs. Of these, 68 new RBP candidates are highly expressed in ESCs compared to differentiated cells, implicating a role in stem-cell physiology. Two well-known E3 ubiquitin ligases, Trim25 (also called Efp) and Trim71 (also called Lin41), are validated as RBPs, revealing a potential link between RNA biology and protein-modification pathways. Our study confirms and expands the atlas of RBPs, providing a useful resource for the study of the RNA-RBP network in stem cells.