Netrin-1-engineered endothelial cell exosomes induce the formation of pre-regenerative niche to accelerate peripheral nerve repair.

The formation of vascular niche is pivotal during the early stage of peripheral nerve regeneration. Nevertheless, the mechanisms of vascular niche in the regulation of peripheral nerve repair remain unclear. Netrin-1 (NTN1) was found up-regulated in nerve stump after peripheral nerve injury (PNI). Herein, we demonstrated that NTN1-high endothelial cells (NTN1+ECs) were the critical component of vascular niche, fostering angiogenesis, axon regeneration, and repair-related phenotypes. We also found that NTN1+EC-derived exosomes (NTN1 EC-EXO) were involved in the formation of vascular niche as a critical role. Multi-omics analysis further verified that NTN1 EC-EXO carried a low-level expression of let7a-5p and activated key pathways associated with niche formation including focal adhesion, axon guidance, phosphatidylinositol 3-kinase-AKT, and mammalian target of rapamycin signaling pathway. Together, our study suggested that the construction of a pre-regenerative niche induced by NTN1 EC-EXO could establish a beneficial microenvironment for nerve repair and facilitate functional recovery after PNI.

Enhancement of Tumor Perfusion and Antiangiogenic Therapy in Murine Models of Clear Cell Renal Cell Carcinoma Using Ultrasound-Stimulated Microbubbles.

OBJECTIVE: To explore the effect of ultrasound-stimulated microbubble cavitation (USMC) on enhancing antiangiogenic therapy in clear cell renal cell carcinoma. MATERIALS AND METHODS: We explored the effects of USMC with different mechanical indices (MIs) on tumor perfusion, 36 786-O tumor-bearing nude mice were randomly assigned into four groups: (i) control group, (ii) USMC0.25 group (MI = 0.25), (iii) USMC1.4 group (MI = 1.4) (iv) US1.4 group (MI = 1.4). Tumor perfusion was assessed by contrast-enhanced ultrasound (CEUS) before the USMC treatment and 30 min, 4h and 6h after the USMC treatment, respectively. Then we evaluated vascular normalization(VN) induced by low-MI (0.25) USMC treatment, 12 tumor-bearing nude mice were randomly divided into two groups: (i) control group (ii) USMC0.25 group. USMC treatment was performed, and tumor microvascular imaging and blood perfusion were analyzed by MicroFlow imaging (MFI) and CEUS 30 min after each treatment. In combination therapy, a total of 144 tumor-bearing nude mice were randomly assigned to six groups (n = 24): (i) control group, (ii) USMC1.4 group, (iii) USMC0.25 group, (iv) bevacizumab(BEV) group, (v) USMC1.4 +BEV group, (vi) USMC0.25 +BEV group. BEV was injected on the 6th, 10th, 14th, and 18th d after the tumors were inoculated, while USMC treatment was performed 24 h before and after every BEV administration. We examined the effects of the combination therapy through a series of experiments. RESULTS: Tumor blood perfusion enhanced by USMC with low MI (0.25)could last for more than 6h, inducing tumor VN and promoting drug delivery. Compared with other groups, USMC0.25+BEV combination therapy had the strongest inhibition on tumor growth, led to the longest survival time of the mice. CONCLUSION: The optimized USMC is a promising therapeutic approach that can be combined with antiangiogenic therapy to combat tumor progression.

Striatum supports fast learning but not memory recall.

Animals learn to carry out motor actions in specific sensory contexts to achieve goals. The striatum has been implicated in producing sensory-motor associations, yet its contribution to memory formation or recall is not clear. To investigate the contribution of striatum to these processes, mice were taught to associate a cue, consisting of optogenetic activation of striatum-projecting neurons in visual cortex, with forelimb reaches to access food pellets. As necessary to direct learning, striatal neural activity encoded both the sensory context and outcome of reaching. With training, the rate of cued reaching increased, but brief optogenetic inhibition of striatal activity arrested learning and prevented trial-to-trial improvements in performance. However, the same manipulation did not affect performance improvements already consolidated into short- (within an hour) or long-term (across days) memories. Hence, striatal activity is necessary for trial-to-trial improvements in task performance, leading to plasticity in other brain areas that mediate memory recall.

Immune response to foreign materials in spinal fusion surgery.

Spinal fusion surgery is a common procedure used to stabilize the spine and treat back pain. The procedure involves the use of foreign materials such as screws, rods, or cages, which can trigger a foreign body reaction, an immune response that involves the activation of immune cells such as macrophages and lymphocytes. The foreign body reaction can impact the success of spinal fusion, as it can interfere with bone growth and fusion. This review article provides an overview of the cellular and molecular events in the foreign body reaction, the impact of the immune response on spinal fusion, and strategies to minimize its impact. By carefully considering the use of foreign materials and optimizing surgical techniques, the impact of the foreign body reaction can be reduced, leading to better outcomes for patients.

Multi-supervised bidirectional fusion network for road-surface condition recognition.

Rapid developments in automatic driving technology have given rise to new experiences for passengers. Safety is a main priority in automatic driving. A strong familiarity with road-surface conditions during the day and night is essential to ensuring driving safety. Existing models used for recognizing road-surface conditions lack the required robustness and generalization abilities. Most studies only validated the performance of these models on daylight images. To address this problem, we propose a novel multi-supervised bidirectional fusion network (MBFN) model to detect weather-induced road-surface conditions on the path of automatic vehicles at both daytime and nighttime. We employed ConvNeXt to extract the basic features, which were further processed using a new bidirectional fusion module to create a fused feature. Then, the basic and fused features were concatenated to generate a refined feature with greater discriminative and generalization abilities. Finally, we designed a multi-supervised loss function to train the MBFN model based on the extracted features. Experiments were conducted using two public datasets. The results clearly demonstrated that the MBFN model could classify diverse road-surface conditions, such as dry, wet, and snowy conditions, with a satisfactory accuracy and outperform state-of-the-art baseline models. Notably, the proposed model has multiple variants that could also achieve competitive performances under different road conditions. The code for the MBFN model is shared at https://zenodo.org/badge/latestdoi/607014079.

Evaluation of the operation status of the residential land market in Beijing-Tianjin-Hebei region of China and its spatiotemporal pattern characteristics.

Exploring the operation status and patterns of urban land markets is an important theoretical and practical topic for promoting coordinated socio-economic development. In this study, the operation status of the residential land market in the Beijing-Tianjin-Hebei region and the characteristics of its pattern were analyzed using the composite index method and the 3σ rule of the normal distribution and taking the 174 counties in Beijing, Tianjin, and Hebei, China, as the research objects. The results show that ① Beijing, Tianjin, Langfang, Zhangjiakou, and Baoding residential land market state composite indexes are all in the middle to upper levels in the Beijing-Tianjin-Hebei city cluster, while Qinhuangdao, Handan, and Chengde residential land market state composite indexes are generally low. The harmony between the residential land price and national economy, the market supply and demand balance, and the structural balance may become the main factors affecting the healthy development of the residential land market in Beijing and Tianjin. ② The proportion of counties with "healthy" residential land market in all dimensions and overall market status reached over 64%, and the residential land market in the Beijing-Tianjin-Hebei region is running well. The rapid increase in residential land prices from 2016 to 2020 was an important driver of the increased heat in the residential land market across the region. ③ The residential land market in the counties around Beijing and Tianjin is significantly hotter than in other regions, and there is an obvious polarization effect in the operation state of the residential land market in the Beijing-Tianjin region. The residential land market is generally cold in the counties in the southern and northeastern parts of the region and other peripheral areas, and there is a risk of marginalization in the development of the residential land market in the counties in the peripheral areas. ④ Both the hot and cold residential land market states exhibit spatial clustering characteristics. Most of the clusters are not consistent with the municipal administrative boundaries, and the states of the residential land market in neighboring counties are very similar.

Thyroid hormone rewires cortical circuits to coordinate body-wide metabolism and exploratory drive.

Animals adapt to varying environmental conditions by modifying the function of their internal organs, including the brain. To be adaptive, alterations in behavior must be coordinated with the functional state of organs throughout the body. Here we find that thyroid hormone- a prominent regulator of metabolism in many peripheral organs- activates cell-type specific transcriptional programs in anterior regions of cortex of adult mice via direct activation of thyroid hormone receptors. These programs are enriched for axon-guidance genes in glutamatergic projection neurons, synaptic regulators across both astrocytes and neurons, and pro-myelination factors in oligodendrocytes, suggesting widespread remodeling of cortical circuits. Indeed, whole-cell electrophysiology recordings revealed that thyroid hormone induces local transcriptional programs that rewire cortical neural circuits via pre-synaptic mechanisms, resulting in increased excitatory drive with a concomitant sensitization of recruited inhibition. We find that thyroid hormone bidirectionally regulates innate exploratory behaviors and that the transcriptionally mediated circuit changes in anterior cortex causally promote exploratory decision-making. Thus, thyroid hormone acts directly on adult cerebral cortex to coordinate exploratory behaviors with whole-body metabolic state.

Dopamine and glutamate regulate striatal acetylcholine in decision-making.

Striatal dopamine and acetylcholine are essential for the selection and reinforcement of motor actions and decision-making1. In vitro studies have revealed an intrastriatal circuit in which acetylcholine, released by cholinergic interneurons (CINs), drives the release of dopamine, and dopamine, in turn, inhibits the activity of CINs through dopamine D2 receptors (D2Rs). Whether and how this circuit contributes to striatal function in vivo is largely unknown. Here, to define the role of this circuit in a living system, we monitored acetylcholine and dopamine signals in the ventrolateral striatum of mice performing a reward-based decision-making task. We establish that dopamine and acetylcholine exhibit multiphasic and anticorrelated transients that are modulated by decision history and reward outcome. Dopamine dynamics and reward encoding do not require the release of acetylcholine by CINs. However, dopamine inhibits acetylcholine transients in a D2R-dependent manner, and loss of this regulation impairs decision-making. To determine how other striatal inputs shape acetylcholine signals, we assessed the contribution of cortical and thalamic projections, and found that glutamate release from both sources is required for acetylcholine release. Altogether, we uncover a dynamic relationship between dopamine and acetylcholine during decision-making, and reveal multiple modes of CIN regulation. These findings deepen our understanding of the neurochemical basis of decision-making and behaviour.

FTO-dependent m6A regulates muscle fiber remodeling in an NFATC1-YTHDF2 dependent manner.

BACKGROUND: Adolescent idiopathic scoliosis (AIS) is characterized by low lean mass without vertebral deformity. The cause-and-effect relationship between scoliosis and paraspinal muscle imbalance has long puzzled researchers. Although FTO has been identified as a susceptibility gene for AIS, its potential role in the asymmetry of paraspinal muscles has not been fully elucidated. METHODS: We investigated the role of Fto in murine myoblast proliferation, migration, and myogenic differentiation. We examined its precise regulatory influence on murine muscle fiber remodeling in vitro and in vivo. We identified the downstream target gene of Fto by screening key regulators of murine muscle fiber remodeling and identified its m6A reader. Deep paraspinal muscle samples were obtained from the concave and convex sides of AIS patients with or without Schroth exercises, and congenital scoliosis served as a control group. We compared the content of type I fibers, expression patterns of fast- and slow-type genes, and levels of FTO expression. RESULTS: FTO contributed to maintain the formation of murine slow-twitch fibers both in vitro and in vivo. These effects were mediated by the demethylation activity of FTO, which specifically demethylated NFATC1 and prevented YTHDF2 from degrading it. We found a significant reduction in type I fibers, mRNA levels of MYH7 and MYH7B, and expression of FTO on the concave side of AIS. The percentage of type I fibers showed a positive correlation with the expression level of FTO. The asymmetric patterns observed in AIS were consistent with those seen in congenital scoliosis, and the asymmetry of FTO expression and fiber type in AIS was largely restored by Schroth exercises. CONCLUSIONS: FTO supports the formation of murine slow-twitch fibers in an NFATC1-YTHDF2 dependent manner. The consistent paraspinal muscle features seen in AIS and congenital scoliosis, as well as the reversible pattern of muscle fibers and expression of FTO in AIS suggest that FTO may contribute to the muscle fiber remodeling secondary to scoliosis.

Impairment of rigidity sensing caused by mutant TP53 gain of function in osteosarcoma.

Osteosarcoma (OS) is the most common primary malignant pediatric bone tumor and is characterized by high heterogeneity. Studies have revealed a wide range of phenotypic differences among OS cell lines in terms of their in vivo tumorigenicity and in vitro colony-forming abilities. However, the underlying molecular mechanism of these discrepancies remains unclear. The potential role of mechanotransduction in tumorigenicity is of particular interest. To this end, we tested the tumorigenicity and anoikis resistance of OS cell lines both in vitro and in vivo. We utilized a sphere culture model, a soft agar assay, and soft and rigid hydrogel surface culture models to investigate the function of rigidity sensing in the tumorigenicity of OS cells. Additionally, we quantified the expression of sensor proteins, including four kinases and seven cytoskeletal proteins, in OS cell lines. The upstream core transcription factors of rigidity-sensing proteins were further investigated. We detected anoikis resistance in transformed OS cells. The mechanosensing function of transformed OS cells was also impaired, with general downregulation of rigidity-sensing components. We identified toggling between normal and transformed growth based on the expression pattern of rigidity-sensing proteins in OS cells. We further uncovered a novel TP53 mutation (R156P) in transformed OS cells, which acquired gain of function to inhibit rigidity sensing, thus sustaining transformed growth. Our findings suggest a fundamental role of rigidity-sensing components in OS tumorigenicity as mechanotransduction elements through which cells can sense their physical microenvironment. In addition, the gain of function of mutant TP53 appears to serve as an executor for such malignant programs.

A novel NSUN5/ENO3 pathway promotes the Warburg effect and cell growth in clear cell renal cell carcinoma by 5-methylcytosine-stabilized ENO3 mRNA.

OBJECTIVES: Clear cell renal cell carcinoma (ccRCC) cells often reprogram their metabolisms. Enolase 3 (ENO3) is closely related to the Warburg effect observed in cells during tumor progression. However, the expression and function of ENO3 in ccRCC cells remain unclear. Therefore, this study investigated the expression and functional significance of ENO3 in the Warburg effect observed in ccRCC cells. METHODS: In this study, B-mode and microflow imaging ultrasound examinations were performed to evaluate patients with ccRCC. The extracellular acidification rate test and glucose uptake and lactate production assays were used to examine the Warburg effect in ccRCC cells. Western blotting, quantitative reverse transcription polymerase chain reaction, and immunochemistry were used to detect the expression of ENO3 and NOP2/Sun RNA methyltransferase 5 (NSUN5). RESULTS: ENO3 upregulation in ccRCC tumor tissues was accompanied by an increase in tumor size. Importantly, ENO3 participated in the Warburg effect observed in ccRCC cells, and high levels of ENO3 indicated a poor prognosis for patients. Loss of ENO3 reduced glucose uptake, lactate production, and extracellular acidification rate as well as inhibited ccRCC cell proliferation. Furthermore, NSUN5 was involved in the ENO3-regulated Warburg effect and ccRCC cell progression. Mechanically, NSUN5 was upregulated in ccRCC tissues, and NSUN5 upregulation mediated 5-methylcytosine modification of messenger RNA (mRNA) in ccRCC cells to promote mRNA stability and ENO3 expression. CONCLUSIONS: Collectively, the destruction of the NSUN5/ENO3 axis prevents ccRCC growth in vivo and in vitro, and targeting this pathway may be an effective strategy against ccRCC progression.

Research on the characteristics and influencing factors of the Beijing-Tianjin-Hebei urban network structure from the perspective of listed manufacturing enterprises.

The production connection is a crucial component of the Inter City Association. An urban network based on the enterprise perspective better reflects the structural characteristics of regional cities. Based on data gathered from the headquarters and branches of the listed manufacturing enterprises in 2020, this paper analyzes the county-level administrative units in the Beijing-Tianjin-Hebei region of China. Using the subordinate connection model and the social network analysis method, this paper examines the spatial structural characteristics and factors of urban networks in the Beijing-Tianjin-Hebei region. The results suggest that resource allocation in the Beijing-Tianjin-Hebei region is unbalanced, with a significant difference in urban radiation and agglomeration capacity. As the administrative centers, Beijing, Tianjin, and Shijiazhuang show a strong ability to allocate resources within the network. The overall network density in the region was shown to be relatively low, with the main links being of low or medium level. The urban network is defined by the network connection led by Beijing and Tianjin with Beijing-Tianjin-Tangshan and Beijing-Baoding-Shijiazhuang as the main axis. These cities exhibit a "dense southeast and sparse northwest" pattern. From a hierarchical perspective, high-level network connections are based mainly on spatial proximity. Analyzing the agglomerative subgroups, the study found that the inner and outer connections of the core subgroups were relatively high. Semi-marginal subgroups generally accepted the radiation of core subgroups, while marginal subgroups had little connection with other subgroups in the region. The results of the QAP analysis show that the administrative relationship, spatial distance, city size difference, economic development level difference, industrial structure difference, and labor cost difference have a significant influence on the urban network of the Beijing-Tianjin-Hebei region.

Neuronal-Activity Dependent Mechanisms of Small Cell Lung Cancer Progression.

Neural activity is increasingly recognized as a critical regulator of cancer growth. In the brain, neuronal activity robustly influences glioma growth both through paracrine mechanisms and through electrochemical integration of malignant cells into neural circuitry via neuron-to-glioma synapses, while perisynaptic neurotransmitter signaling drives breast cancer brain metastasis growth. Outside of the CNS, innervation of tumors such as prostate, breast, pancreatic and gastrointestinal cancers by peripheral nerves similarly regulates cancer progression. However, the extent to which the nervous system regulates lung cancer progression, either in the lung or when metastatic to brain, is largely unexplored. Small cell lung cancer (SCLC) is a lethal high-grade neuroendocrine tumor that exhibits a strong propensity to metastasize to the brain. Here we demonstrate that, similar to glioma, metastatic SCLC cells in the brain co-opt neuronal activity-regulated mechanisms to stimulate growth and progression. Optogenetic stimulation of cortical neuronal activity drives proliferation and invasion of SCLC brain metastases. In the brain, SCLC cells exhibit electrical currents and consequent calcium transients in response to neuronal activity, and direct SCLC cell membrane depolarization is sufficient to promote the growth of SCLC tumors. In the lung, vagus nerve transection markedly inhibits primary lung tumor formation, progression and metastasis, highlighting a critical role for innervation in overall SCLC initiation and progression. Taken together, these studies illustrate that neuronal activity plays a crucial role in dictating SCLC pathogenesis in both primary and metastatic sites.

Endothelial cell-derived exosomes boost and maintain repair-related phenotypes of Schwann cells via miR199-5p to promote nerve regeneration.

BACKGROUND: Schwann cells (SCs) respond to nerve injury by transforming into the repair-related cell phenotype, which can provide the essential signals and spatial cues to promote axonal regeneration and induce target reinnervation. Endothelial cells (ECs) contribute to intraneural angiogenesis contributing to creating a permissive microenvironment. The coordination between ECs and SCs within injury sites is crucial in the regeneration process, however, it still unclear. As the intercellular vital information mediators in the nervous system, exosomes have been proposed to take a significant role in regulating regeneration. Thus, the main purpose of this study is to determine the facilitative effect of ECs-derived exosomes on SCs and to seek the underlying mechanism. RESULTS: In the present study, we collected exosomes from media of ECs. We demonstrated that exosomes derived from ECs possessed the favorable neuronal affinity both in vitro and in vivo. Further research indicated that EC-exosomes (EC-EXO) could boost and maintain repair-related phenotypes of SCs, thereby enhancing axonal regeneration, myelination of regenerated axons and neurologically functional recovery of the injured nerve. MiRNA sequencing in EXO-treated SCs and control SCs indicated that EC-EXO significantly up-regulated expression of miR199-5p. Furthermore, this study demonstrated that EC-EXO drove the conversion of SC phenotypes in a PI3K/AKT/PTEN-dependent manner. CONCLUSION: In conclusion, our research indicates that the internalization of EC-EXO in SCs can promote nerve regeneration by boosting and maintaining the repair-related phenotypes of SCs. And the mechanism may be relevant to the up-regulated expression of miR199-5p and activation of PI3K/AKT/PTEN signaling pathway.

Predicting miRNA-disease associations based on lncRNA-miRNA interactions and graph convolution networks.

Increasing studies have proved that microRNAs (miRNAs) are critical biomarkers in the development of human complex diseases. Identifying disease-related miRNAs is beneficial to disease prevention, diagnosis and remedy. Based on the assumption that similar miRNAs tend to associate with similar diseases, various computational methods have been developed to predict novel miRNA-disease associations (MDAs). However, selecting proper features for similarity calculation is a challenging task because of data deficiencies in biomedical science. In this study, we propose a deep learning-based computational method named MAGCN to predict potential MDAs without using any similarity measurements. Our method predicts novel MDAs based on known lncRNA-miRNA interactions via graph convolution networks with multichannel attention mechanism and convolutional neural network combiner. Extensive experiments show that the average area under the receiver operating characteristic values obtained by our method under 2-fold, 5-fold and 10-fold cross-validations are 0.8994, 0.9032 and 0.9044, respectively. When compared with five state-of-the-art methods, MAGCN shows improvement in terms of prediction accuracy. In addition, we conduct case studies on three diseases to discover their related miRNAs, and find that all the top 50 predictions for all the three diseases have been supported by established databases. The comprehensive results demonstrate that our method is a reliable tool in detecting new disease-related miRNAs.

Preparation of ultrasound contrast agents: The exploration of the structure-echogenicity relationship of contrast agents based on neural network model.

There is a need to standardize the process of micro/nanobubble preparation to bring it closer to clinical translation. We explored a neural network-based model to predict the structure-echogenicity relationship for the preparation and fabrication of ultrasound-enhanced contrast agents. Seven formulations were screened, and 109 measurements were obtained. An artificial neural network-multilayer perceptron (ANN-MLP) model was used. The original data were divided into the training and testing groups, which included 73 and 36 groups of data, respectively. The hidden layer was selected from three hidden layers and included bias. The classification graph showed that the predicted values of the training and testing groups were 76.7% and 66.7%, respectively. According to the receiver operating characteristic curve, the accuracy of different imaging effects could achieve a prediction rate of 88.1-96.5%. The percentage graph showed that the data were gradually converging. The predictive analysis curves of different ultrasound effects gradually approached stable value of Gain. Normalized importance predicted contributions for the Pk1, poly-dispersity index (PDI), and intensity account were 100%, 98.5%, and 89.7%, respectively. The application of the ANN-MLP model is feasible and effective for the exploration of the synthesis process of ultrasound contrast agents. 1,2-Distearoyl-sn-glycero-3 phosphoethanolamine-N (methoxy[polyethylene glycol]-2000) (DSPE PEG-2000) correlated highly with the success rate of contrast agent synthesis.

Predicting potential miRNA-disease associations based on more reliable negative sample selection.

BACKGROUND: Increasing biomedical studies have shown that the dysfunction of miRNAs is closely related with many human diseases. Identifying disease-associated miRNAs would contribute to the understanding of pathological mechanisms of diseases. Supervised learning-based computational methods have continuously been developed for miRNA-disease association predictions. Negative samples of experimentally-validated uncorrelated miRNA-disease pairs are required for these approaches, while they are not available due to lack of biomedical research interest. Existing methods mainly choose negative samples from the unlabelled ones randomly. Therefore, the selection of more reliable negative samples is of great importance for these methods to achieve satisfactory prediction results. RESULTS: In this study, we propose a computational method termed as KR-NSSM which integrates two semi-supervised algorithms to select more reliable negative samples for miRNA-disease association predictions. Our method uses a refined K-means algorithm for preliminary screening of likely negative and positive miRNA-disease samples. A Rocchio classification-based method is applied for further screening to receive more reliable negative and positive samples. We implement ablation tests in KR-NSSM and find that the combination of the two selection procedures would obtain more reliable negative samples for miRNA-disease association predictions. Comprehensive experiments based on fivefold cross-validations demonstrate improvements in prediction accuracy on six classic classifiers and five known miRNA-disease association prediction models when using negative samples chose by our method than by previous negative sample selection strategies. Moreover, 469 out of 1123 selected positive miRNA-disease associations by our method are confirmed by existing databases. CONCLUSIONS: Our experiments show that KR-NSSM can screen out more reliable negative samples from the unlabelled ones, which greatly improves the performance of supervised machine learning methods in miRNA-disease association predictions. We expect that KR-NSSM would be a useful tool in negative sample selection in biomedical research.

Predicting miRNA-disease associations based on graph attention networks and dual Laplacian regularized least squares.

Increasing biomedical evidence has proved that the dysregulation of miRNAs is associated with human complex diseases. Identification of disease-related miRNAs is of great importance for disease prevention, diagnosis and remedy. To reduce the time and cost of biomedical experiments, there is a strong incentive to develop efficient computational methods to infer potential miRNA-disease associations. Although many computational approaches have been proposed to address this issue, the prediction accuracy needs to be further improved. In this study, we present a computational framework MKGAT to predict possible associations between miRNAs and diseases through graph attention networks (GATs) using dual Laplacian regularized least squares. We use GATs to learn embeddings of miRNAs and diseases on each layer from initial input features of known miRNA-disease associations, intra-miRNA similarities and intra-disease similarities. We then calculate kernel matrices of miRNAs and diseases based on Gaussian interaction profile (GIP) with the learned embeddings. We further fuse the kernel matrices of each layer and initial similarities with attention mechanism. Dual Laplacian regularized least squares are finally applied for new miRNA-disease association predictions with the fused miRNA and disease kernels. Compared with six state-of-the-art methods by 5-fold cross-validations, our method MKGAT receives the highest AUROC value of 0.9627 and AUPR value of 0.7372. We use MKGAT to predict related miRNAs for three cancers and discover that all the top 50 predicted results in the three diseases are confirmed by existing databases. The excellent performance indicates that MKGAT would be a useful computational tool for revealing disease-related miRNAs.

RASSF8-AS1 displays low expression in colorectal cancer and up-regulates RASSF8 to suppress cell invasion and migration.

BACKGROUND: Colorectal cancer (CRC) is among the most prevalent cancers. Long non-coding RNAs (lncRNAs) are important participant in various cancers. Based on the literature, lncRNA RASSF8-AS1 inhibits laryngeal squamous cell carcinoma (LSCC) malignant progression. However, the role of RASSF8-AS1 in CRC remains unclear. PURPOSE: This study centered on uncovering the role of RASSF8-AS1 and its related regulatory mechanisms in CRC cells. METHODS: RT-qPCR and western blot were performed to examine the expression of target genes. Functional assays were conducted to determine the effect of target genes on the migration and invasion of CRC cells. Mechanism assays were also carried out to figure out the specific downstream mechanisms of RASSF8-AS1. In vivo assays were also involved. RESULTS: The expression of RASSF8-AS1 and RASSF8 was positively correlated in CRC, and the two genes were down-regulated in CRC cells and tissues. Moreover, CRC cell invasion and migration as well as xenograft CRC tumor growth suppressed by RASSF8-AS1 overexpression were entirely recovered by RASSF8 knockdown or partially rescued by miR-33a-5p augment. As for the downstream mechanism, RASSF8-AS1 sponged miR-33a-5p to up-regulate RASSF8, or recruited HNRNPC to stabilize RASSF8 mRNA. CONCLUSION: RASSF8-AS1 modulates miR-33a-5p/HNRNPC/RASSF8 axis to further impede CRC cell invasion and migration. AVAILABILITY OF DATA: The research data is confidential.