A tripartite rheostat controls self-regulated host plant resistance to insects.

Plants deploy receptor-like kinases and nucleotide-binding leucine-rich repeat receptors to confer host plant resistance (HPR) to herbivores1. These gene-for-gene interactions between insects and their hosts have been proposed for more than 50 years2. However, the molecular and cellular mechanisms that underlie HPR have been elusive, as the identity and sensing mechanisms of insect avirulence effectors have remained unknown. Here we identify an insect salivary protein perceived by a plant immune receptor. The BPH14-interacting salivary protein (BISP) from the brown planthopper (Nilaparvata lugens Stål) is secreted into rice (Oryza sativa) during feeding. In susceptible plants, BISP targets O. satvia RLCK185 (OsRLCK185; hereafter Os is used to denote O. satvia-related proteins or genes) to suppress basal defences. In resistant plants, the nucleotide-binding leucine-rich repeat receptor BPH14 directly binds BISP to activate HPR. Constitutive activation of Bph14-mediated immunity is detrimental to plant growth and productivity. The fine-tuning of Bph14-mediated HPR is achieved through direct binding of BISP and BPH14 to the selective autophagy cargo receptor OsNBR1, which delivers BISP to OsATG8 for degradation. Autophagy therefore controls BISP levels. In Bph14 plants, autophagy restores cellular homeostasis by downregulating HPR when feeding by brown planthoppers ceases. We identify an insect saliva protein sensed by a plant immune receptor and discover a three-way interaction system that offers opportunities for developing high-yield, insect-resistant crops.

FHL3 promotes the formation of fast glycolytic muscle fibers by interacting with YY1 and muscle glycolytic metabolism.

The proportions of the various muscle fiber types are important in the regulation of skeletal muscle metabolism, as well as animal meat production. Four-and-a-half LIM domain protein 3 (FHL3) is highly expressed in fast glycolytic muscle fibers and differentially regulates the expression of myosin heavy chain (MyHC) isoforms at the cellular level. Whether FHL3 regulates the transformation of muscle fiber types in vivo and the regulatory mechanism is unclear. In this study, muscle-specific FHL3 transgenic mice were generated by random integration, and lentivirus-mediated gene knockdown or overexpression in muscles of mice or pigs was conducted. Functional analysis showed that overexpression of FHL3 in muscles significantly increased the proportion of fast-twitch myofibers and muscle mass but decreased muscle succinate dehydrogenase (SDH) activity and whole-body oxygen consumption. Lentivirus-mediated FHL3 knockdown in muscles significantly decreased muscle mass and the proportion of fast-twitch myofibers. Mechanistically, FHL3 directly interacted with the Yin yang 1 (YY1) DNA-binding domain, repressed the binding of YY1 to the fast glycolytic MyHC2b gene regulatory region, and thereby promoted MyHC2b expression. FHL3 also competed with EZH2 to bind the repression domain of YY1 and reduced H3K27me3 enrichment in the MyHC2b regulatory region. Moreover, FHL3 overexpression reduced glucose tolerance by affecting muscle glycolytic metabolism, and its mRNA expression in muscle was positively associated with hemoglobin A1c (HbA1c) in patients with type 2 diabetes. Therefore, FHL3 is a novel potential target gene for the treatment of muscle metabolism-related diseases and improvement of animal meat production.

Functional SNPs in SYISL promoter significantly affect muscle fiber density and muscle traits in pigs.

Our previous studies showed that SYISL is a negative regulator of muscle growth and regeneration in mice, pigs and humans. SYISL knockout resulted in an increase in the density of muscle fibers and muscle growth. However, it is unclear whether there are natural mutations in pig SYNPO2 intron sense-overlapping lncRNA (pSYISL) that affect the expression of pSYISL and muscle growth traits. In this study, three SNPs in exons and six SNPs within the promoter of pSYISL were identified. Association analysis showed that the two SNPs in exons are significantly associated with loin muscle area (p < 0.05); the six SNPs in the promoter that show complete linkage are significantly associated with live backfat thickness and live loin muscle area in American Large White pigs. Bioinformatics and luciferase reporter assays as well as in vitro binding experiments indicated that the mutation of SNP rs702045770 (g.539G>A) leads to the loss of YY1 binding to the promoter, thus affecting the expression level of pSYISL, and we found that Jiangshan Black pigs with genotype GG have a higher expression level of pSYISL than genotype AA individuals, but the muscle fiber density was significantly lower than in genotype AA individuals. Furthermore, the association analysis showed that the carcass backfat thickness of genotype GG of SNP rs702045770 was significantly higher than that of other genotypes in (Pietrain × Duroc) × (Landrace × Yorkshire) crossbred pigs (p < 0.05). The glycolytic potential of genotype GG was significantly higher than that of other genotypes (p < 0.05). These results provide novel insight into the identification of functional SNPs in non-coding genomic regions.

Long noncoding RNA lncMREF promotes myogenic differentiation and muscle regeneration by interacting with the Smarca5/p300 complex.

Long noncoding RNAs (lncRNAs) play important roles in the spatial and temporal regulation of muscle development and regeneration. Nevertheless, the determination of their biological functions and mechanisms underlying muscle regeneration remains challenging. Here, we identified a lncRNA named lncMREF (lncRNA muscle regeneration enhancement factor) as a conserved positive regulator of muscle regeneration among mice, pigs and humans. Functional studies demonstrated that lncMREF, which is mainly expressed in differentiated muscle satellite cells, promotes myogenic differentiation and muscle regeneration. Mechanistically, lncMREF interacts with Smarca5 to promote chromatin accessibility when muscle satellite cells are activated and start to differentiate, thereby facilitating genomic binding of p300/CBP/H3K27ac to upregulate the expression of myogenic regulators, such as MyoD and cell differentiation. Our results unravel a novel temporal-specific epigenetic regulation during muscle regeneration and reveal that lncMREF/Smarca5-mediated epigenetic programming is responsible for muscle cell differentiation, which provides new insights into the regulatory mechanism of muscle regeneration.

Targeted overexpression of PPARγ in skeletal muscle by random insertion and CRISPR/Cas9 transgenic pig cloning enhances oxidative fiber formation and intramuscular fat deposition.

Peroxisome proliferator-activated receptor gamma (PPARγ) is a master regulator of adipogenesis and lipogenesis. To understand its roles in fiber formation and fat deposition in skeletal muscle, we successfully generated muscle-specific overexpression of PPARγ in two pig models by random insertion and CRISPR/Cas9 transgenic cloning procedures. The content of intramuscular fat was significantly increased in PPARγ pigs while had no changes on lean meat ratio. PPARγ could promote adipocyte differentiation by activating adipocyte differentiating regulators such as FABP4 and CCAAT/enhancer-binding protein (C/EBP), along with enhanced expression of LPL, FABP4, and PLIN1 to proceed fat deposition. Proteomics analyses demonstrated that oxidative metabolism of fatty acids and respiratory chain were activated in PPARγ pigs, thus, gathered more Ca2+ in PPARγ pigs. Raising of Ca2+ could result in increased phosphorylation of CAMKII and p38 MAPK in PPARγ pigs, which can stimulate MEF2 and PGC1α to affect fiber type and oxidative capacity. These results support that skeletal muscle-specific overexpression of PPARγ can promote oxidative fiber formation and intramuscular fat deposition in pigs.

Large-Scale Place Recognition Based on Camera-LiDAR Fused Descriptor.

In the field of autonomous driving, carriers are equipped with a variety of sensors, including cameras and LiDARs. However, the camera suffers from problems of illumination and occlusion, and the LiDAR encounters motion distortion, degenerate environment and limited ranging distance. Therefore, fusing the information from these two sensors deserves to be explored. In this paper, we propose a fusion network which robustly captures both the image and point cloud descriptors to solve the place recognition problem. Our contribution can be summarized as: (1) applying the trimmed strategy in the point cloud global feature aggregation to improve the recognition performance, (2) building a compact fusion framework which captures both the robust representation of the image and 3D point cloud, and (3) learning a proper metric to describe the similarity of our fused global feature. The experiments on KITTI and KAIST datasets show that the proposed fused descriptor is more robust and discriminative than the single sensor descriptor.

Efficient Robust Watermarking Based on Structure-Preserving Quaternion Singular Value Decomposition.

Quaternion singular value decomposition (QSVD) is a robust technique of digital watermarking that extracts high quality watermarks from watermarked images with low distortion. However, the existing QSVD-based watermarking schemes face the obstacle of "explosion of complexity" and have much room for improvement in terms of real-time, invisibility, and robustness. In this paper, we overcome such obstacle by introducing a new real structure-preserving QSVD algorithm and propose a novel QSVD-based watermarking scheme with high efficiency. Secret information is transmitted blindly by incorporating two new strategies: coefficient pair selection and adaptive embedding. The highly correlated coefficient pairs determined by the normalized cross-correlation method reduce the impact of embedding by reducing the maximum modification of the coefficient values, resulting in high fidelity of the watermarked image. Large-size 8-color binary watermark and QR code effectively verify that the proposed watermarking scheme can resist various image attacks in numerical experiments. Two keys designed by Logistic chaotic map ensure the security of the watermarking system. Under the premise of considering the correlation of color channels, the proposed watermarking scheme not only performs well in real-time and invisibility, but also has satisfactory advantages in robustness compared with the state-of-the-art methods.

Distributional generative adversarial imitation learning with reproducing kernel generalization.

Generative adversarial imitation learning (GAIL) regards imitation learning (IL) as a distribution matching problem between the state-action distributions of the expert policy and the learned policy. In this paper, we focus on the generalization and computational properties of policy classes. We prove that the generalization can be guaranteed in GAIL when the class of policies is well controlled. With the capability of policy generalization, we introduce distributional reinforcement learning (RL) into GAIL and propose the greedy distributional soft gradient (GDSG) algorithm to solve GAIL. The main advantages of GDSG can be summarized as: (1) Q-value overestimation, a crucial factor leading to the instability of GAIL with off-policy training, can be alleviated by distributional RL. (2) By considering the maximum entropy objective, the policy can be improved in terms of performance and sample efficiency through sufficient exploration. Moreover, GDSG attains a sublinear convergence rate to a stationary solution. Comprehensive experimental verification in MuJoCo environments shows that GDSG can mimic expert demonstrations better than previous GAIL variants.

SAB Net: A Semantic Attention Boosting Framework for Semantic Segmentation.

Semantic segmentation has achieved great progress by effectively fusing features of contextual information. In this article, we propose an end-to-end semantic attention boosting (SAB) framework to adaptively fuse the contextual information iteratively across layers with semantic regularization. Specifically, we first propose a pixelwise semantic attention (SAP) block, with a semantic metric representing the pixelwise category relationship, to aggregate the nonlocal contextual information. In addition, we improve the computation complexity of SAP block from O(n²) to O(n) for images with size n. Second, we present a categorywise semantic attention (SAC) block to adaptively balance the nonlocal contextual dependencies and the local consistency with a categorywise weight, overcoming the contextual information confusion caused by the feature imbalance within intra-category. Furthermore, we propose the SAB module to refine the segmentation with SAC and SAP blocks. By applying the SAB module iteratively across layers, our model shrinks the semantic gap and enhances the structure reasoning by fully utilizing the coarse segmentation information. Extensive quantitative evaluations demonstrate that our method significantly improves the segmentation results and achieves superior performance on the PASCAL VOC 2012, Cityscapes, PASCAL Context, and ADE20K datasets.

Conservative analysis of Synaptopodin-2 intron sense-overlapping lncRNA reveals its novel function in promoting muscle atrophy.

BACKGROUND: Dissection of the regulatory pathways that control skeletal muscle development and atrophy is important for the treatment of muscle wasting. Long noncoding RNA (lncRNA) play important roles in various stages of muscle development. We previously reported that Synaptopodin-2 (SYNPO2) intron sense-overlapping lncRNA (SYISL) regulates myogenesis through an interaction with enhancer of zeste homologue 2 (EZH2). However, it remains unclear whether SYISL homologues exist in humans and pigs, and whether the functions and mechanisms of these homologues are conserved among species. METHODS: Bioinformatics, cell fractionation, and quantitative real-time polymerase chain reaction (qRT-PCR) analyses were used for the identification and molecular characterization of SYISL homologues in humans and pigs. Effects on myogenesis and muscle atrophy were determined via loss-of-function or gain-of-function experiments using C2C12 myoblasts, myogenic progenitor cells, dexamethasone (DEX), and aging-induced muscle atrophy models. RNA pulldown, RNA immunoprecipitation, dual luciferase reporting, and co-transfection experiments were used to explore the mechanisms of SYISL interactions with proteins and miRNAs. RESULTS: We identified SYISL homologues in humans (designated hSYISL) and pigs (designated pSYISL). Functional experiments demonstrated that hSYISL and pSYISL regulate myogenesis through interactions with EZH2. Interestingly, we showed that SYISL functions to regulate muscle atrophy and sarcopenia through comparative analysis. SYISL is significantly up-regulated after muscle atrophy (P < 0.01); it significantly promotes muscle atrophy in DEX-induced muscle atrophy models (P < 0.01). SYISL knockdown or knockout alleviates muscle atrophy and sarcopenia in DEX-induced and aged mice. The tibialis anterior (TA) muscle weight of 3-month-old wild-type (WT) mice decreased by 33.24% after DEX treatment (P < 0.001), while the muscle weight loss of 3-month-old SYISL knockout mice was only 18.20% after DEX treatment (P < 0.001). SYISL knockout in 18-month-old WT mice significantly increased the weights of quadriceps (Qu), gastrocnemius (Gas), and TA muscles by 10.45% (P < 0.05), 13.95% (P < 0.01), and 24.82% (P < 0.05), respectively. Mechanistically, SYISL increases the expression levels of the muscle atrophy genes forkhead box protein O3a (FoxO3a), muscle ring finger 1 (MuRF1), and muscle atrophy-related F-box (Atrogin-1) via sponging of miR-23a-3p/miR-103-3p/miR-205-5p and thus promotes muscle atrophy. Additionally, we verified that human SYISL overexpression in muscles of 18-month-old WT mice significantly decreased the weights of Gas, Qu, and TA muscles by 7.76% (P < 0.01), 12.26% (P < 0.05), and 13.44% (P < 0.01), respectively, and accelerates muscle atrophy through conserved mechanisms. CONCLUSIONS: Our results identify SYISL as a conserved lncRNA that modulates myogenesis in mice, pigs, and humans. We also demonstrated its previously unknown ability to promote muscle atrophy.

Single Nucleotide Polymorphisms of Porcine lncMGPF Regulate Meat Production Traits by Affecting RNA Stability.

lncMGPF is a novel positive regulator of myogenic differentiation, muscle growth and regeneration in mouse, pig, and human. But whether natural mutations within lncMGPF gene regulate animal meat production traits is unclear. In this study, ten single nucleotide polymorphisms (SNPs) of pig lncMGPF (plncMGPF) gene were identified among commercial pig breeds and Chinese local pig breeds. These SNPs are highly linked and constructed into multiple haplotypes, and haplotype ATTCATGTTC (H1) mainly exists in commercial pig breeds while haplotype GCCTGCACCT (H3) is more frequent in Chinese local pig breeds. Association analysis indicated that all SNPs are significantly associated with the backfat thickness and loin muscle area (P < 0.05), respectively, and homologous H1 individuals have higher loin muscle area and lower backfat thickness than H3 pigs. Bioinformatics and functional analysis showed that haplotype H1 has a longer half-life and more stable RNA secondary structure than haplotype H3. plncMGPF haplotype H1 has stronger effects on pig primary myogenic progenitor cells differentiation and muscle growth than haplotype H3. Further experiments showed that two SNPs (rs81403974 and rs325492834) function together to confer plncMGPF stability and function. Our observation suggested that the SNPs in lncMGPF can change the RNA stabilities and lncMGPF function, thereby affecting the porcine meat production traits.

Lipidomic analyses reveal enhanced lipolysis in planthoppers feeding on resistant host plants.

The brown planthopper (BPH) (Nilaparvata lugens Stål) is a highly destructive pest that seriously damages rice (Oryza sativa L.) and causes severe yield losses. To better understand the physiological and metabolic mechanisms through which BPHs respond to resistant rice, we combined mass-spectrometry-based lipidomics with transcriptomic analysis and gene knockdown techniques to compare the lipidomes of BPHs feeding on either of the two resistant (NIL-Bph6 and NIL-Bph9) plants or a wild-type, BPH susceptible (9311) plant. Insects that were fed on resistant rice transformed triglyceride (TG) to phosphatidylcholine (PC) and digalactosyldiacylglycerol (DGDG), with these lipid classes showing significant alterations in fatty acid composition. Moreover, the insects that were fed on resistant rice were characterized by prominent expression changes in genes involved in lipid metabolism processes. Knockdown of the NlBmm gene, which encodes a lipase that regulates the mobilization of lipid reserves, significantly increased TG content and feeding performance of BPHs on resistant plants relative to dsGFP-injected BPHs. Our study provides the first detailed description of lipid changes in BPHs fed on resistant and susceptible rice genotypes. Results from BPHs fed on resistant rice plants reveal that these insects can accelerate TG mobilization to provide energy for cell proliferation, body maintenance, growth and oviposition.

CAB U-Net: An end-to-end category attention boosting algorithm for segmentation.

With the development of machine learning and artificial intelligence, many convolutional neural networks (CNNs) based segmentation methods have been proposed for 3D cardiac segmentation. In this paper, we propose the category attention boosting (CAB) module, which combines the deep network calculation graph with the boosting method. On the one hand, we add the attention mechanism into the gradient boosting process, which enhances the information of coarse segmentation without high computation cost. On the other hand, we introduce the CAB module into the 3D U-Net segmentation network and construct a new multi-scale boosting model CAB U-Net which strengthens the gradient flow in the network and makes full use of the low resolution feature information. Thanks to the advantage that end-to-end networks can adaptively adjust the internal parameters, CAB U-Net can make full use of the complementary effects among different base learners. Extensive experiments on public datasets show that our approach can achieve superior performance over the state-of-the-art methods.

Salivary Protein 1 of Brown Planthopper Is Required for Survival and Induces Immunity Response in Plants.

The brown planthopper (BPH), Nilaparvata lugens Stål, is one of the major pests of rice. It uses its stylet to penetrate rice phloem, feeding on rice sap and causing direct damage to rice or even plant death. During the feeding process, BPHs secrete saliva into plant tissues, which plays crucial roles in the plant-insect interactions. However, little is known about how the salivary proteins secreted by BPH affect feeding ability and how they induce plant immune responses. Here, we identified an N. lugens Salivary Protein 1 (NlSP1) by screening salivary proteome and characterized its functions in BPH and plants. NlSP1 induces cell death, H2O2 accumulation, the expression of defense-related genes, and callose deposition in planta. The active region of NlSP1 that induces plant cell death is located in its N-terminal region. Inhibition of NlSP1 expression in BPHs reduced their feeding ability and had a lethal effect on them. Most importantly, we demonstrated that NlSP1 was able to be secreted into rice plant during feeding process and form a complex with certain interacting partner of rice. These results provide a detailed characterization of a salivary protein from BPHs and offers new insights into our understanding of rice-BPH interaction.

Nonlinear Semi-Supervised Metric Learning Via Multiple Kernels and Local Topology.

Changing the metric on the data may change the data distribution, hence a good distance metric can promote the performance of learning algorithm. In this paper, we address the semi-supervised distance metric learning (ML) problem to obtain the best nonlinear metric for the data. First, we describe the nonlinear metric by the multiple kernel representation. By this approach, we project the data into a high dimensional space, where the data can be well represented by linear ML. Then, we reformulate the linear ML by a minimization problem on the positive definite matrix group. Finally, we develop a two-step algorithm for solving this model and design an intrinsic steepest descent algorithm to learn the positive definite metric matrix. Experimental results validate that our proposed method is effective and outperforms several state-of-the-art ML methods.

Manifold Preserving: An Intrinsic Approach for Semisupervised Distance Metric Learning.

In this paper, we address the semisupervised distance metric learning problem and its applications in classification and image retrieval. First, we formulate a semisupervised distance metric learning model by considering the metric information of inner classes and interclasses. In this model, an adaptive parameter is designed to balance the inner metrics and intermetrics by using data structure. Second, we convert the model to a minimization problem whose variable is symmetric positive-definite matrix. Third, in implementation, we deduce an intrinsic steepest descent method, which assures that the metric matrix is strictly symmetric positive-definite at each iteration, with the manifold structure of the symmetric positive-definite matrix manifold. Finally, we test the proposed algorithm on conventional data sets, and compare it with other four representative methods. The numerical results validate that the proposed method significantly improves the classification with the same computational efficiency.

Nonlinear image registration with bidirectional metric and reciprocal regularization.

Nonlinear registration is an important technique to align two different images and widely applied in medical image analysis. In this paper, we develop a novel nonlinear registration framework based on the diffeomorphic demons, where a reciprocal regularizer is introduced to assume that the deformation between two images is an exact diffeomorphism. In detail, first, we adopt a bidirectional metric to improve the symmetry of the energy functional, whose variables are two reciprocal deformations. Secondly, we slack these two deformations into two independent variables and introduce a reciprocal regularizer to assure the deformations being the exact diffeomorphism. Then, we utilize an alternating iterative strategy to decouple the model into two minimizing subproblems, where a new closed form for the approximate velocity of deformation is calculated. Finally, we compare our proposed algorithm on two data sets of real brain MR images with two relative and conventional methods. The results validate that our proposed method improves accuracy and robustness of registration, as well as the gained bidirectional deformations are actually reciprocal.