DNA Transformer for Visualizing Endogenous RNA Dynamics in Live Cells

Fluorescent DNA tetrahedral probe with catalytic hairpin self-assembly reaction for imaging of miR-21 and miR-155 in living cells

A CHA-based fluorescent DNA tetrahedral probe (FDTp) has been designed to detect the microRNAs miR-21 and miR-155 sensitively and specifically in living cells. The design consisted of functional elements (H1, H2, and Protector) connected to a DNA tetrahedron modified with two pairs of fluorophores and quenching groups. In the presence of miR-21, the chain displacement effect was triggered and Cy3 fluorescence was emitted. In the presence of miR-155, the signal of the catalytic hairpin assembly (CHA) between H1 and H2 on FDTp was amplified, making the fluorescence of FAM sensitive to miR-155. Using this method, the detection limit for miR-155 was 5 pM. The FDTp successfully imaged miR-21 and miR-155 in living cells and distinguished a variety of cell lines based on their expression levels of miR-21 and miR-155. The detection and imaging of dual targets in this design ensured the accuracy of tumor diagnosis and provided a new method for early tumor diagnosis.

Long noncoding RNA MALAT1 is dynamically regulated in leader cells during collective cancer invasion

Cancer cells collectively invade using a leader-follower organization, but the regulation of leader cells during this dynamic process is poorly understood. Using a dual double-stranded locked nucleic acid (LNA) nanobiosensor that tracks long noncoding RNA (lncRNA) dynamics in live single cells, we monitored the spatiotemporal distribution of lncRNA during collective cancer invasion. We show that the lncRNA MALAT1 (metastasis-associated lung adenocarcinoma transcript 1) is dynamically regulated in the invading fronts of cancer cells and patient-derived spheroids. MALAT1 transcripts exhibit distinct abundance, diffusivity, and distribution between leader and follower cells. MALAT1 expression increases when a cancer cell becomes a leader and decreases when the collective migration process stops. Transient knockdown of MALAT1 prevents the formation of leader cells and abolishes the invasion of cancer cells. Taken together, our single-cell analysis suggests that MALAT1 is dynamically regulated in leader cells during collective cancer invasion.

Oncogene-targeting nanoprobes for early imaging detection of tumor

Malignant tumors have been one of the major reasons for deaths worldwide. Timely and accurate diagnosis as well as effective intervention of tumors play an essential role in the survival of patients. Genomic instability is the important foundation and feature of cancer, hence, in vivo oncogene imaging based on novel probes provides a valuable tool for the diagnosis of cancer at early-stage. However, the in vivo oncogene imaging is confronted with great challenge, due to the extremely low copies of oncogene in tumor cells. By combining with various novel activatable probes, the molecular imaging technologies provide a feasible approach to visualize oncogene in situ, and realize accurate treatment of tumor. This review aims to declare the design of nanoprobes responded to tumor associated DNA or RNA, and summarize their applications in detection and bioimaging for tumors. The significant challenges and prospective of oncogene-targeting nanoprobes towards tumors diagnosis are revealed as well.

Shedding Light on DNA-Based Nanoprobes for Live-Cell MicroRNA Imaging

DNA-based nanoprobes integrated with various imaging signals have been employed for fabricating versatile biosensor platforms for the study of intracellular biological process and biomarker detection. The nanoprobes developments also provide opportunities for endogenous microRNA (miRNA) in situ analysis. In this review, the authors are primarily interested in various DNA-based nanoprobes for miRNA biosensors and declare strategies to reveal how to customize the desired nanoplatforms. Initially, various delivery vehicles for nanoprobe architectures transmembrane transport are delineated, and their biosecurity and ability for resisting the complex cellular environment are evaluated. Then, the novel strategies for designing DNA sequences as target miRNA specific recognition and signal amplification modules for miRNA detection are presented. Afterward, recent advances in imaging technologies to accurately respond and produce significant signal output are summarized. Finally, the challenges and future directions in the field are discussed.

miRe2e: a full end-to-end deep model based on transformers for prediction of pre-miRNAs

MOTIVATION

MicroRNAs (miRNAs) are small RNA sequences with key roles in the regulation of gene expression at post-transcriptional level in different species. Accurate prediction of novel miRNAs is needed due to their importance in many biological processes and their associations with complicated diseases in humans. Many machine learning approaches were proposed in the last decade for this purpose, but requiring handcrafted features extraction to identify possible de novo miRNAs. More recently, the emergence of deep learning (DL) has allowed the automatic feature extraction, learning relevant representations by themselves. However, the state-of-art deep models require complex pre-processing of the input sequences and prediction of their secondary structure to reach an acceptable performance.

RESULTS

In this work, we present miRe2e, the first full end-to-end DL model for pre-miRNA prediction. This model is based on Transformers, a neural architecture that uses attention mechanisms to infer global dependencies between inputs and outputs. It is capable of receiving the raw genome-wide data as input, without any pre-processing nor feature engineering. After a training stage with known pre-miRNAs, hairpin and non-harpin sequences, it can identify all the pre-miRNA sequences within a genome. The model has been validated through several experimental setups using the human genome, and it was compared with state-of-the-art algorithms obtaining 10 times better performance.

AVAILABILITY AND IMPLEMENTATION

Webdemo available at https://sinc.unl.edu.ar/web-demo/miRe2e/ and source code available for download at https://github.com/sinc-lab/miRe2e.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Optimizing locked nucleic acid modification in double-stranded biosensors for live single cell analysis

Double-stranded (ds) biosensors are homogeneous oligonucleotide probes for detection of nucleic acid sequences in biochemical assays and live cell imaging.

Intratumoral Heterogeneity Promotes Collective Cancer Invasion through NOTCH1 Variation

Cellular and molecular heterogeneity within tumors has long been associated with the progression of cancer to an aggressive phenotype and a poor prognosis. However, how such intratumoral heterogeneity contributes to the invasiveness of cancer is largely unknown. Here, using a tumor bioengineering approach, we investigate the interaction between molecular subtypes within bladder microtumors and the corresponding effects on their invasiveness. Our results reveal heterogeneous microtumors formed by multiple molecular subtypes possess enhanced invasiveness compared to individual cells, even when both cells are not invasive individually. To examine the molecular mechanism of intratumoral heterogeneity mediated invasiveness, live single cell biosensing, RNA interference, and CRISPR-Cas9 gene editing approaches were applied to investigate and control the composition of the microtumors. An agent-based computational model was also developed to evaluate the influence of NOTCH1 variation on DLL4 expression within a microtumor. The data indicate that intratumoral variation in NOTCH1 expression can lead to upregulation of DLL4 expression within the microtumor and enhancement of microtumor invasiveness. Overall, our results reveal a novel mechanism of heterogeneity mediated invasiveness through intratumoral variation of gene expression.

Decoding leader cells in collective cancer invasion

Collective cancer invasion with leader-follower organization is increasingly recognized as a predominant mechanism in the metastatic cascade. Leader cells support cancer invasion by creating invasion tracks, sensing environmental cues and coordinating with follower cells biochemically and biomechanically. With the latest developments in experimental and computational models and analysis techniques, the range of specific traits and features of leader cells reported in the literature is rapidly expanding. Yet, despite their importance, there is no consensus on how leader cells arise or their essential characteristics. In this Perspective, we propose a framework for defining the essential aspects of leader cells and provide a unifying perspective on the varying cellular and molecular programmes that are adopted by each leader cell subtype to accomplish their functions. This Perspective can lead to more effective strategies to interdict a major contributor to metastatic capability.

Self-Assembled Growing DNA Tree Mediated by Exosomes for Amplified Imaging of Messenger RNA in Living Cells

Sensitive, accurate, and nondestructive probing of endogenous messenger RNA (mRNA) in living cells places extremely high demands on nanocarriers and probes and is still a challenge. In the present study, we describe a target-triggered self-assembled DNA tree for amplified analysis of mRNA in intact living cells. The probes assembled into a DNA tree are transported into cells by exosomes, which is beneficial for reducing cell damage and realizing nondestructive analysis. The probes are l-configured single-stranded DNAs (LDNAs) that can resist the degradation of exonuclease and endonuclease, thus laying the foundation for accurate analysis. Under the induction of the target mRNA, the probes in the cells assemble into a small plantlet and eventually grow into a tree after a few rounds of self-cycling, achieving the exponential amplification of fluorescence signals. Compared with the signal amplification based on one-dimensional DNA trunk self-assembly, the three-dimensional DNA tree shows an excellent sensitivity both ex situ and in situ. In this way, favorable sensitivity, accuracy, and nondestructive analysis are integrated into one system. This DNA tree expands the analysis platform for analyzing more biomarkers on a genetic level in an intracellular, nondestructive, and hypersensitive manner and holds great potential in clinical diagnostic and research applications.

DNA nanostructure-based nucleic acid probes: construction and biological applications

In recent years, DNA has been widely noted as a kind of material that can be used to construct building blocks for biosensing, in vivo imaging, drug development, and disease therapy because of its advantages of good biocompatibility and programmable properties. However, traditional DNA-based sensing processes are mostly achieved by random diffusion of free DNA probes, which were restricted by limited dynamics and relatively low efficiency. Moreover, in the application of biosystems, single-stranded DNA probes face challenges such as being difficult to internalize into cells and being easily decomposed in the cellular microenvironment. To overcome the above limitations, DNA nanostructure-based probes have attracted intense attention. This kind of probe showed a series of advantages compared to the conventional ones, including increased biostability, enhanced cell internalization efficiency, accelerated reaction rate, and amplified signal output, and thus improved in vitro and in vivo applications. Therefore, reviewing and summarizing the important roles of DNA nanostructures in improving biosensor design is very necessary for the development of DNA nanotechnology and its applications in biology and pharmacology. In this perspective, DNA nanostructure-based probes are reviewed and summarized from several aspects: probe classification according to the dimensions of DNA nanostructures (one, two, and three-dimensional nanostructures), the common connection modes between nucleic acid probes and DNA nanostructures, and the most important advantages of DNA self-assembled nanostructures in the applications of biosensing, imaging analysis, cell assembly, cell capture, and theranostics. Finally, the challenges and prospects for the future development of DNA nanostructure-based nucleic acid probes are also discussed.

Zippering DNA Tetrahedral Hyperlink for Ultrasensitive Electrochemical MicroRNA Detection

Pluripotency of a DNA tetrahedron (DNA^TT) has made the iconic framework a compelling keystone in biosensors and biodevices. Herein, distinct from the well-tapped applications in substrate fabrication, we focus on exploring their tracing and signaling potentials. A homologous family of four isostructural DNA^TT, i.e., DNA^{TTα/β/γ/δ}, was engineered to form a sensor circuitry, in which a target-specific monolayer of thiolated DNA^TTγ pinned down the analyte jointly with the reciprocal DNA^TTδ into a sandwich complex; the latter further rallied an in situ interdigital relay of biotinylated DNA^TTα/β into a microsized hyperlink dubbed polyDNA^TT. Its scale and growth factors were illuminated rudimentarily in transmission electron microscopy and confocal laser scanning microscopy. Using a nonsmall-cell lung cancer-related microRNA (hsa-miR-193a-3p) as the subject, a compound DNA-backboned construct was synthesized, fusing all building blocks together. Its superb tacticity and stereochemical conformality avail the templating of a horseradish peroxidase train, which boosted the paralleled catalytic surge of proton donors, resulting in an attomolar detection limit and a broad calibration range of more than seven orders of magnitude. Such oligomerization bested the conventional hybridization chain reaction laddering at both biomechanical stability and stoichiometric congruency. More significantly, it demonstrates the flexibility of DNA architectures and their multitasking ability in biosensing.

Dynamic DNA Assemblies in Biomedical Applications

Deoxyribonucleic acid (DNA) has been widely used to construct homogeneous structures with increasing complexity for biological and biomedical applications due to their powerful functionalities. Especially, dynamic DNA assemblies (DDAs) have demonstrated the ability to simulate molecular motions and fluctuations in bionic systems. DDAs, including DNA robots, DNA probes, DNA nanochannels, DNA templates, etc., can perform structural transformations or predictable behaviors in response to corresponding stimuli and show potential in the fields of single molecule sensing, drug delivery, molecular assembly, etc. A wave of exploration of the principles in designing and usage of DDAs has occurred, however, knowledge on these concepts is still limited. Although some previous reviews have been reported, systematic and detailed reviews are rare. To achieve a better understanding of the mechanisms in DDAs, herein, the recent progress on the fundamental principles regarding DDAs and their applications are summarized. The relative assembly principles and computer-aided software for their designing are introduced. The advantages and disadvantages of each software are discussed. The motional mechanisms of the DDAs are classified into exogenous and endogenous stimuli-triggered responses. The special dynamic behaviors of DDAs in biomedical applications are also summarized. Moreover, the current challenges and future directions of DDAs are proposed.

Subcellular Spatial Transcriptomes: Emerging Frontier for Understanding Gene Regulation

RNAs are trafficked and localized with exquisite precision inside the cell. Studies of candidate messenger RNAs have shown the vital importance of RNA subcellular location in development and cellular function. New sequencing- and imaging-based methods are providing complementary insights into subcellular localization of RNAs transcriptome-wide. APEX-seq and ribosome profiling as well as proximity-labeling approaches have revealed thousands of transcript isoforms are localized to distinct cytotopic locations, including locations that defy biochemical fractionation and hence were missed by prior studies. Sequences in the 3' and 5' untranslated regions (UTRs) serve as "zip codes" to direct transcripts to particular locales, and it is clear that intronic and retrotransposable sequences within transcripts have been co-opted by cells to control localization. Molecular motors, nuclear-to-cytosol RNA export, liquid-liquid phase separation, RNA modifications, and RNA structure dynamically shape the subcellular transcriptome. Location-based RNA regulation continues to pose new mysteries for the field, yet promises to reveal insights into fundamental cell biology and disease mechanisms.

Intracellular Entropy-Driven Multi-Bit DNA Computing for Tumor Progression Discrimination

Tumor progressions such as metastasis are complicated events that involve abnormal expression of different miRNAs and enzymes. Monitoring these biomolecules in live cells with computational DNA nanotechnology may enable discrimination of tumor progression via digital outputs. Herein, we report intracellular entropy-driven multivalent DNA circuits to implement multi-bit computing for simultaneous analysis of intracellular telomerase and microRNAs including miR-21 and miR-31. These three biomolecules can trigger respective DNA strand displacement recycling reactions for signal amplification. They are visualized by fluorescence imaging, and their signal outputs are encoded as multi-bit binary codes for different cell types. The results can discriminate non-tumorigenic, malignant and metastatic breast cells as well as respective tumors. This DNA computing circuit is further performed in a microfluidic chip to differentiate rare co-cultured cells, which holds a potential for the analysis of clinical samples.

One DNA circle capture probe with multiple target recognition domains for simultaneous electrochemical detection of miRNA-21 and miRNA-155

In this work, a novel DNA circle capture probe with multiple target recognition domains was designed to develop an electrochemical biosensor for ultrasensitive detection of microRNA-21 (miRNA-21) and miRNA-155 simultaneously. The DNA circle capture probe was anchored at the top of the tetrahedron DNA nanostructure (TDN) to simultaneously recognize miRNA-21 and miRNA-155 through multiple target recognition domains under the assistance of Helper strands, which could trigger mimetic proximity ligation assay (mPLA) for capturing the beacons ferrocene (Fc)-A1 and methylene blue (MB)-A2 to achieve multiple miRNAs detection. In this way, the local reaction concentration could be enhanced and avoid the interference of various capture probes compared with the traditional multiplexed electrochemical biosensor with the use of different capture probes, resulting in the significantly improvement of detection sensitivity. As a result, this proposed biosensor showed wide linearity ranging from 0.1 fM to 10 nM with detection limits of miRNA-21 and miRNA-155 as 18.9 aM and 39.6 aM respectively, which also could be applied in the simultaneously detection of miRNA-21 and miRNA-155 from cancer cell lysates. The present strategy paved a new path in the design of capture probes for achieving more efficient and sensitive multiple biomarkers detections and possessed the potential applications in clinical diagnostic of diseases.

Electrocatalytic Efficiency Regulation between Target-Induced HRP-Mimicking DNAzyme and GOx with Low Background for Ultrasensitive Detection of Thrombin

Herein, an efficient target-activated enzyme cascade electrocatalysis with low background signal was employed to establish electrochemical biosensor for ultrasensitive detection of thrombin via regulating electrocatalytic efficiency between target-induced hemin/G-quadruplex horseradish peroxidase-mimicking DNAzyme (HRP-mimicking DNAzyme) and glucose oxidase (GOx). Impressively, only when the target thrombin was introduced, the HRP-mimicking DNAzyme acting simultaneously as electrochemical signal probe would be formed to activate high-efficiency enzyme cascade electrocatalysis for reducing background signal significantly, which could overcome the defect of inevitable high background signal during the detection of target in the traditional cascade electrocatalysis of two existing bioenzymes. In addition, the detection sensitivity could be further improved by regulating the side length of rigid DNA tetrahedron (TDN) scaffold anchored HRP-mimicking DNAzyme and GOx at adjacent vertices for high enzyme cascade electrocatalytic efficiency. Consequently, the proposed biosensor demonstrated a low detection limit down to 0.3 fM for target thrombin, which provided a promising method for ultrasensitive monitoring of biomolecules in sensing analysis and disease diagnosis.