Towards regulatory generative AI in ophthalmology healthcare: a security and privacy perspective.

As the healthcare community increasingly harnesses the power of generative artificial intelligence (AI), critical issues of security, privacy and regulation take centre stage. In this paper, we explore the security and privacy risks of generative AI from model-level and data-level perspectives. Moreover, we elucidate the potential consequences and case studies within the domain of ophthalmology. Model-level risks include knowledge leakage from the model and model safety under AI-specific attacks, while data-level risks involve unauthorised data collection and data accuracy concerns. Within the healthcare context, these risks can bear severe consequences, encompassing potential breaches of sensitive information, violating privacy rights and threats to patient safety. This paper not only highlights these challenges but also elucidates governance-driven solutions that adhere to AI and healthcare regulations. We advocate for preparedness against potential threats, call for transparency enhancements and underscore the necessity of clinical validation before real-world implementation. The objective of security and privacy improvement in generative AI warrants emphasising the role of ophthalmologists and other healthcare providers, and the timely introduction of comprehensive regulations.

MultiFair: Model Fairness With Multiple Sensitive Attributes.

While existing fairness interventions show promise in mitigating biased predictions, most studies concentrate on single-attribute protections. Although a few methods consider multiple attributes, they either require additional constraints or prediction heads, incurring high computational overhead or jeopardizing the stability of the training process. More critically, they consider per-attribute protection approaches, raising concerns about fairness gerrymandering where certain attribute combinations remain unfair. This work aims to construct a neutral domain containing fused information across all subgroups and attributes. It delivers fair predictions as the fused input contains neutralized information for all considered attributes. Specifically, we adopt mixup operations to generate samples with fused information. However, our experiments reveal that directly adopting the operations leads to degraded prediction results. The excessive mixup operations result in unrecognizable training data. To this end, we design three distinct mixup schemes that balance information fusion across attributes while retaining distinct visual features critical for training valid models. Extensive experiments with multiple datasets and up to eight sensitive attributes demonstrate that the proposed MultiFair method can deliver fairness protections for multiple attributes while maintaining valid prediction results.

Myeloid BAF60a deficiency alters metabolic homeostasis and exacerbates atherosclerosis.

Atherosclerosis, a leading health concern, stems from the dynamic involvement of immune cells in vascular plaques. Despite its significance, the interplay between chromatin remodeling and transcriptional regulation in plaque macrophages is understudied. We discovered the reduced expression of Baf60a, a component of the switch/sucrose non-fermentable (SWI/SNF) chromatin remodeling complex, in macrophages from advanced plaques. Myeloid-specific Baf60a deletion compromised mitochondrial integrity and heightened adhesion, apoptosis, and plaque development. BAF60a preserves mitochondrial energy homeostasis under pro-atherogenic stimuli by retaining nuclear respiratory factor 1 (NRF1) accessibility at critical genes. Overexpression of BAF60a rescued mitochondrial dysfunction in an NRF1-dependent manner. This study illuminates the BAF60a-NRF1 axis as a mitochondrial function modulator in atherosclerosis, proposing the rejuvenation of perturbed chromatin remodeling machinery as a potential therapeutic target.

Analysis of polycomb repressive complex 2 (PRC2) subunits in Picea abies with a focus on embryo development.

BACKGROUND: Conserved polycomb repressive complex 2 (PRC2) mediates H3K27me3 to direct transcriptional repression and has a key role in cell fate determination and cell differentiation in both animals and plants. PRC2 subunits have undergone independent multiplication and functional divergence in higher plants. However, relevant information is still absent in gymnosperms. RESULTS: To launch gymnosperm PRC2 research, we identified and cloned the PRC2 core component genes in the conifer model species Picea abies, including one Esc/FIE homolog PaFIE, two p55/MSI homologs PaMSI1a and PaMSI1b, two E(z) homologs PaKMT6A2 and PaKMT6A4, a Su(z)12 homolog PaEMF2 and a PaEMF2-like fragment. Phylogenetic and protein domain analyses were conducted. The Esc/FIE homologs were highly conserved in the land plant, except the monocots. The other gymnospermous PRC2 subunits underwent independent evolution with angiospermous species to different extents. The relative transcript levels of these genes were measured in endosperm and zygotic and somatic embryos at different developmental stages. The obtained results proposed the involvement of PaMSI1b and PaKMT6A4 in embryogenesis and PaKMT6A2 and PaEMF2 in the transition from embryos to seedlings. The PaEMF2-like fragment was predominantly expressed in the endosperm but not in the embryo. In addition, immunohistochemistry assay showed that H3K27me3 deposits were generally enriched at meristem regions during seed development in P. abies. CONCLUSIONS: This study reports the first characterization of the PRC2 core component genes in the coniferous species P. abies. Our work may enable a deeper understanding of the cell reprogramming process during seed and embryo development and may guide further research on embryonic potential and development in conifers.

Generating endogenous Myh11-driven Cre mice for sex-independent gene deletion in smooth muscle cells.

Specific and efficient smooth muscle cell-targeted (SMC-targeted) gene deletion is typically achieved by pairing SMMHC-CreERT2-Tg mice with mice carrying the loxP-flanked gene. However, the transgene, CreERT2, is not controlled by the endogenous Myh11 gene promoter, and the codon-modified iCreERT2 exhibits significant tamoxifen-independent leakage. Furthermore, because the Cre-bearing bacterial artificial chromosome (BAC) is inserted onto the Y chromosome, the SMMHC-CreERT2-Tg mice strain can only exhibit gene deletions in male mice. Additionally, there is a lack of Myh11-driven constitutive Cre mice when tamoxifen usage is a concern. We used CRISPR/Cas9-mediated homologous recombination between a donor vector carrying the CreNLSP2A or CreERT2-P2A sequence and homologous arm surrounding the translation start site of the Myh11 gene to generate Cre-knockin mice. The P2A sequence enables the simultaneous translation of Cre and endogenous proteins. Using reporter mice, we assessed Cre-mediated recombination efficiency, specificity, tamoxifen-dependent controllability, and functionality in both sexes. Both constitutive (Myh11-CreNLSP2A) and inducible (Myh11-CreERT2-P2A) Cre mice demonstrated efficient, SMC-specific, sex-independent Cre recombinase activity without confounding endogenous gene expression. Combined with recently generated BAC transgenic Myh11-CreERT2-RAD mice and the Itga8-CreERT2 mouse models, our models will help expand the research toolbox, facilitating unbiased and comprehensive research in SMCs and SMC-dependent cardiovascular diseases.

Balancing Learning Model Privacy, Fairness, and Accuracy With Early Stopping Criteria.

As deep learning models mature, one of the most prescient questions we face is: what is the ideal tradeoff between accuracy, fairness, and privacy (AFP)? Unfortunately, both the privacy and the fairness of a model come at the cost of its accuracy. Hence, an efficient and effective means of fine-tuning the balance between this trinity of needs is critical. Motivated by some curious observations in privacy-accuracy tradeoffs with differentially private stochastic gradient descent (DP-SGD), where fair models sometimes result, we conjecture that fairness might be better managed as an indirect byproduct of this process. Hence, we conduct a series of analyses, both theoretical and empirical, on the impacts of implementing DP-SGD in deep neural network models through gradient clipping and noise addition. The results show that, in deep learning, the number of training epochs is central to striking a balance between AFP because DP-SGD makes the training less stable, providing the possibility of model updates at a low discrimination level without much loss in accuracy. Based on this observation, we designed two different early stopping criteria to help analysts choose the optimal epoch at which to stop training a model so as to achieve their ideal tradeoff. Extensive experiments show that our methods can achieve an ideal balance between AFP.

Model-Based Self-Advising for Multi-Agent Learning.

In multiagent learning, one of the main ways to improve learning performance is to ask for advice from another agent. Contemporary advising methods share a common limitation that a teacher agent can only advise a student agent if the teacher has experience with an identical state. However, in highly complex learning scenarios, such as autonomous driving, it is rare for two agents to experience exactly the same state, which makes the advice less of a learning aid and more of a one-time instruction. In these scenarios, with contemporary methods, agents do not really help each other learn, and the main outcome of their back and forth requests for advice is an exorbitant communications' overhead. In human interactions, teachers are often asked for advice on what to do in situations that students are personally unfamiliar with. In these, we generally draw from similar experiences to formulate advice. This inspired us to provide agents with the same ability when asked for advice on an unfamiliar state. Hence, we propose a model-based self-advising method that allows agents to train a model based on states similar to the state in question to inform its response. As a result, the advice given can not only be used to resolve the current dilemma but also many other similar situations that the student may come across in the future via self-advising. Compared with contemporary methods, our method brings a significant improvement in learning performance with much lower communication overheads.

Defense against membership inference attack in graph neural networks through graph perturbation.

Graph neural networks have demonstrated remarkable performance in learning node or graph representations for various graph-related tasks. However, learning with graph data or its embedded representations may induce privacy issues when the node representations contain sensitive or private user information. Although many machine learning models or techniques have been proposed for privacy preservation of traditional non-graph structured data, there is limited work to address graph privacy concerns. In this paper, we investigate the privacy problem of embedding representations of nodes, in which an adversary can infer the user's privacy by designing an inference attack algorithm. To address this problem, we develop a defense algorithm against white-box membership inference attacks, based on perturbation injection on the graph. In particular, we employ a graph reconstruction model and inject a certain size of noise into the intermediate output of the model, i.e., the latent representations of the nodes. The experimental results obtained on real-world datasets, along with reasonable usability and privacy metrics, demonstrate that our proposed approach can effectively resist membership inference attacks. Meanwhile, based on our method, the trade-off between usability and privacy brought by defense measures can be observed intuitively, which provides a reference for subsequent research in the field of graph privacy protection.

BAF60c prevents abdominal aortic aneurysm formation through epigenetic control of vascular smooth muscle cell homeostasis.

Abdominal aortic aneurysm (AAA) is a life-threatening vascular disease. BAF60c, a unique subunit of the SWItch/sucrose nonfermentable (SWI/SNF) chromatin remodeling complex, is critical for cardiac and skeletal myogenesis, yet little is known about its function in the vasculature and, specifically, in AAA pathogenesis. Here, we found that BAF60c was downregulated in human and mouse AAA tissues, with primary staining to vascular smooth muscle cells (VSMCs), confirmed by single-cell RNA-sequencing. In vivo studies revealed that VSMC-specific knockout of Baf60c significantly aggravated both angiotensin II- (Ang II-) and elastase-induced AAA formation in mice, with a significant increase in elastin degradation, inflammatory cell infiltration, VSMC phenotypic switch, and apoptosis. In vitro studies showed that BAF60c knockdown in VSMCs resulted in loss of contractile phenotype, increased VSMC inflammation, and apoptosis. Mechanistically, we demonstrated that BAF60c preserved VSMC contractile phenotype by strengthening serum response factor (SRF) association with its coactivator P300 and the SWI/SNF complex and suppressing VSMC inflammation by promoting a repressive chromatin state of NF-κB target genes as well as preventing VSMC apoptosis through transcriptional activation of KLF5-dependent B cell lymphoma 2 (BCL2) expression. Our identification of the essential role of BAF60c in preserving VSMC homeostasis expands its therapeutic potential in preventing and treating AAA.

Integrated Lipidomic and Transcriptomic Analysis Reveals Phospholipid Changes in Somatic Embryos of Picea asperata in Response to Partial Desiccation.

Partial desiccation treatment (PDT) is an effective technology for promoting the germination and conversion of conifer somatic embryos (SEs). PDT, as a drought stress, induces intensive physiological responses in phospholipid metabolism, which are not well understood in the conifer SEs. Here, we integrated lipidomics, transcriptomics and proteomics analyses to reveal the molecular basis of lipid remodeling under PDT in Picea asperata SEs. Among the 82 lipid molecular species determined by mass spectrometry, phosphatidic acid (PA) had a significant effect after PDT and was the most critical lipid in the response to PDT. The transcriptomics results showed that multiple transcripts in the glycerolipid and glycerophospholipid metabolism pathways were differentially expressed, and these included five PLDα1 transcripts that catalyze the conversion of phosphatidylcholine (PC) to PA. Furthermore, the enzyme activity of this phospholipase D (PLD) was significantly enhanced in response to PDT, and PDT also significantly increased the protein level of PLDα1 (MA_10436582g0020). In addition, PA is a key factor in gibberellin, abscisic acid and ethylene signal transduction. One GDI1, one DELLA, three ABI1s, two SnRK2s, one CTR and 12 ERFs showed significantly differential expression between SEs before and after PDT in this study. Our data suggest that the observed increases in the PA contents might result from the activation of PLDα by PDT. PA not only affects the physical and chemical properties of the cell membrane but also participates in plant hormone signal transduction. Our work provides novel insight into the molecular mechanism through which PDT promotes the germination of SEs of coniferous tree species and fills the gap in the understanding of the mechanism of somatic embryo lipid remodeling in response to PDT.

Differential Advising in Multiagent Reinforcement Learning.

Agent advising is one of the main approaches to improve agent learning performance by enabling agents to share advice. Existing advising methods have a common limitation that an adviser agent can offer advice to an advisee agent only if the advice is created in the same state as the advisee's state. However, in complex environments, it is a very strong requirement that two states are the same, because a state may consist of multiple dimensions and two states being the same means that all these dimensions in the two states are correspondingly identical. Therefore, this requirement may limit the applicability of existing advising methods to complex environments. In this article, inspired by the differential privacy scheme, we propose a differential advising method that relaxes this requirement by enabling agents to use advice in a state even if the advice is created in a slightly different state. Compared with the existing methods, agents using the proposed method have more opportunity to take advice from others. This article is the first to adopt the concept of differential privacy on advising to improve agent learning performance instead of addressing security issues. The experimental results demonstrate that the proposed method is more efficient in complex environments than the existing methods.

Transcriptomic, Proteomic, and Metabolic Profiles of Catalpa bungei Tension Wood Reveal New Insight Into Lignin Biosynthesis Involving Transcription Factor Regulation.

Lignin is a complex polymer in plant cell walls whose proportion is second only to that of cellulose and plays an important role in the mechanical properties of wood and stress resistance of plants. Here, we induced tension wood (TW) formation in Catalpa bungei by artificial bending and analyzed the lignin metabolism of the TW. LC-MS analysis showed that a significantly higher content of coniferyl aldehyde was observed in the TW cell wall than in the opposite wood (OW) and normal wood (NW) cell walls. TW had significantly lower contents of coniferyl alcohol than OW and NW. Raman spectroscopy results indicated that TW had lower total lignin than OW and NW. The transcription and translation levels of most of the differentially expressed genes (DEGs) involved in lignin monomer biosynthesis indicated upregulation in TW/OW and TW/NW. We found no significant difference in the transcription levels of three collision gases (CADs) between TW and OW or between NW, but their translation levels were significantly downregulated in TW, suggesting post-transcriptional control for CAD. We predicted and analyzed transcription factors that could target DEGs involved in lignin monomer biosynthesis in TW. Based on the analysis of the relationships of targeting and coexpression, we found that NAC (evm.model.group1.695) could potentially target 4CLs and CCoAOMT, that HD-Zip (evm.model.group7.1157) had potential targeting relationships with CCoAOMT, F5H, and CCR, and that their expression levels were significantly positive. It is speculated that the upregulation of NAC and HD-ZIP transcription factors activates the expression of downstream target genes, which leads to a significant increase in coniferyl aldehyde in TW. However, the decrease in total lignin in TW may be caused by the significant downregulation of CAD translation and the significant decrease in precursors (coniferyl alcohol). Whether the expression of CAD genes is regulated by post-transcriptional control and affects TW lignin metabolism needs further study.

A Bivariate Mapping Model Identifies Major Covariation QTLs for Biomass Allocation Between Leaf and Stem Growth of Catalpa bungei.

Biomass allocation plays a critical role in plant morphological formation and phenotypic plasticity, which greatly impact plant adaptability and competitiveness. While empirical studies on plant biomass allocation have focused on molecular biology and ecology approaches, detailed insight into the genetic basis of biomass allocation between leaf and stem growth is still lacking. Herein, we constructed a bivariate mapping model to identify covariation QTLs governing carbon (C) allocation between the leaves and stem as well as the covariation of traits within and between organs in a full-sib mapping population of C. bungei. A total of 123 covQTLs were detected for 23 trait pairs, including six leaf traits (leaf length, width, area, perimeter, length/width ratio and petiole length) and five stem traits (height, diameter at breast height, wood density, stemwood volume and stemwood biomass). The candidate genes were further identified in tissue-specific gene expression data, which provided insights into the genetic architecture underlying C allocation for traits or organs. The key QTLs related to growth and biomass allocation, which included UVH1, CLPT2, GAD/SPL, COG1 and MTERF4, were characterised and verified via gene function annotation and expression profiling. The integration of a bivariate Quantitative trait locus mapping model and gene expression profiling will enable the elucidation of genetic architecture underlying biomass allocation and covariation growth, in turn providing a theoretical basis for forest molecular marker-assisted breeding with specific C allocation strategies for adaptation to heterogeneous environments.

Single-Cell Transcriptomics Reveals Endothelial Plasticity During Diabetic Atherogenesis.

Atherosclerosis is the leading cause of cardiovascular diseases, which is also the primary cause of mortality among diabetic patients. Endothelial cell (EC) dysfunction is a critical early step in the development of atherosclerosis and aggravated in the presence of concurrent diabetes. Although the heterogeneity of the organ-specific ECs has been systematically analyzed at the single-cell level in healthy conditions, their transcriptomic changes in diabetic atherosclerosis remain largely unexplored. Here, we carried out a single-cell RNA sequencing (scRNA-seq) study using EC-enriched single cells from mouse heart and aorta after 12 weeks feeding of a standard chow or a diabetogenic high-fat diet with cholesterol. We identified eight EC clusters, three of which expressed mesenchymal markers, indicative of an endothelial-to-mesenchymal transition (EndMT). Analyses of the marker genes, pathways, and biological functions revealed that ECs are highly heterogeneous and plastic both in normal and atherosclerotic conditions. The metabolic transcriptomic analysis further confirmed that EndMT-derived fibroblast-like cells are prominent in atherosclerosis, with diminished fatty acid oxidation and enhanced biological functions, including regulation of extracellular-matrix organization and apoptosis. In summary, our data characterized the phenotypic and metabolic heterogeneity of ECs in diabetes-associated atherogenesis at the single-cell level and paves the way for a deeper understanding of endothelial cell biology and EC-related cardiovascular diseases.

KLF11 protects against abdominal aortic aneurysm through inhibition of endothelial cell dysfunction.

Abdominal aortic aneurysm (AAA) is a life-threatening degenerative vascular disease. Endothelial cell (EC) dysfunction is implicated in AAA. Our group recently demonstrated that Krüppel-like factor 11 (KLF11) plays an essential role in maintaining vascular homeostasis, at least partially through inhibition of EC inflammatory activation. However, the functions of endothelial KLF11 in AAA remain unknown. Here we found that endothelial KLF11 expression was reduced in the ECs from human aneurysms and was time dependently decreased in the aneurysmal endothelium from both elastase- and Pcsk9/AngII-induced AAA mouse models. KLF11 deficiency in ECs markedly aggravated AAA formation, whereas EC-selective KLF11 overexpression markedly inhibited AAA formation. Mechanistically, KLF11 not only inhibited the EC inflammatory response but also diminished MMP9 expression and activity and reduced NADPH oxidase 2-mediated production of reactive oxygen species in ECs. In addition, KLF11-deficient ECs induced smooth muscle cell dedifferentiation and apoptosis. Overall, we established endothelial KLF11 as a potentially novel factor protecting against AAA and a potential target for intervention in aortic aneurysms.

Human apolipoprotein A-II reduces atherosclerosis in knock-in rabbits.

BACKGROUND AND AIMS: Apolipoprotein A-II (apoAII) is the second major apolipoprotein of the high-density lipoprotein (HDL) particle, after apoAI. Unlike apoAI, the biological and physiological functions of apoAII are unclear. We aimed to gain insight into the specific roles of apoAII in lipoprotein metabolism and atherosclerosis using a novel rabbit model. METHODS: Wild-type (WT) rabbits are naturally deficient in apoAII, thus their HDL contains only apoAI. Using TALEN technology, we replaced the endogenous apoAI in rabbits through knock-in (KI) of human apoAII. The newly generated apoAII KI rabbits were used to study the specific function of apoAII, independent of apoAI. RESULTS: ApoAII KI rabbits expressed exclusively apoAII without apoAI, as confirmed by RT-PCR and Western blotting. On a standard diet, the KI rabbits exhibited lower plasma triglycerides (TG, 52%, p < 0.01) due to accelerated clearance of TG-rich particles and higher lipoprotein lipase activity than the WT littermates. ApoAII KI rabbits also had higher plasma HDL-C (28%, p < 0.05) and their HDL was rich in apoE, apoAIV, and apoAV. When fed a cholesterol-rich diet for 16 weeks, apoAII KI rabbits were resistant to diet-induced hypertriglyceridemia and developed significantly less aortic atherosclerosis compared to WT rabbits. HDL isolated from rabbits with apoAII KI had similar cholesterol efflux capacity and anti-inflammatory effects as HDL isolated from the WT rabbits. CONCLUSIONS: ApoAII KI rabbits developed less atherosclerosis than WT rabbits, possibly through increased plasma HDL-C, reduced TG and atherogenic lipoproteins. These results suggest that apoAII may serve as a potential target for the treatment of atherosclerosis.

Suppression of Vascular Macrophage Activation by Nitro-Oleic Acid and its Implication for Abdominal Aortic Aneurysm Therapy.

PURPOSE: Abdominal aortic aneurysm (AAA) is one of the leading causes of death in the developed world and is currently undertreated due to the complicated nature of the disease. Herein, we aimed to address the therapeutic potential of a novel class of pleiotropic mediators, specifically a new drug candidate, nitro-oleic acid (NO2-OA), on AAA, in a well-characterized murine AAA model. METHODS: We generated AAA using a mouse model combining AAV.PCSK9-D377Y induced hypercholesterolemia with angiotensin II given by chronic infusion. Vehicle control (PEG-400), oleic acid (OA), or NO2-OA were subcutaneously delivered to mice using an osmotic minipump. We characterized the effects of NO2-OA on pathophysiological responses and dissected the underlying molecular mechanisms through various in vitro and ex vivo strategies. RESULTS: Subcutaneous administration of NO2-OA significantly decreased the AAA incidence (8/28 mice) and supra-renal aorta diameters compared to mice infused with either PEG-400 (13/19, p = 0.0117) or OA (16/23, p = 0.0078). In parallel, the infusion of NO2-OA in the AAA model drastically decreased extracellular matrix degradation, inflammatory cytokine levels, and leucocyte/macrophage infiltration in the vasculature. Administration of NO2-OA reduced inflammation, cytokine secretion, and cell migration triggered by various biological stimuli in primary and macrophage cell lines partially through activation of the peroxisome proliferator-activated receptor-gamma (PPARγ). Moreover, the protective effect of NO2-OA relies on the inhibition of macrophage prostaglandin E2 (PGE2)-induced PGE2 receptor 4 (EP4) cAMP signaling, known to participate in the development of AAA. CONCLUSION: Administration of NO2-OA protects against AAA formation and multifactorial macrophage activation. With NO2-OA currently undergoing FDA approved phase II clinical trials, these findings may expedite the use of this nitro-fatty acid for AAA therapy.

Single-cell RNA sequencing reveals the cellular heterogeneity of aneurysmal infrarenal abdominal aorta.

AIMS: The artery contains numerous cell types which contribute to multiple vascular diseases. However, the heterogeneity and cellular responses of these vascular cells during abdominal aortic aneurysm (AAA) progression have not been well characterized. METHODS AND RESULTS: Single-cell RNA sequencing was performed on the infrarenal abdominal aortas (IAAs) from C57BL/6J mice at Days 7 and 14 post-sham or peri-adventitial elastase-induced AAA. Unbiased clustering analysis of the transcriptional profiles from >4500 aortic cells identified 17 clusters representing nine-cell lineages, encompassing vascular smooth muscle cells (VSMCs), fibroblasts, endothelial cells, immune cells (macrophages, T cells, B cells, and dendritic cells), and two types of rare cells, including neural cells and erythrocyte cells. Seurat clustering analysis identified four smooth muscle cell (SMC) subpopulations and five monocyte/macrophage subpopulations, with distinct transcriptional profiles. During AAA progression, three major SMC subpopulations were proportionally decreased, whereas the small subpopulation was increased, accompanied with down-regulation of SMC contractile markers and up-regulation of pro-inflammatory genes. Another AAA-associated cellular response is immune cell expansion, particularly monocytes/macrophages. Elastase exposure induced significant expansion and activation of aortic resident macrophages, blood-derived monocytes and inflammatory macrophages. We also identified increased blood-derived reparative macrophages expressing anti-inflammatory cytokines suggesting that resolution of inflammation and vascular repair also persist during AAA progression. CONCLUSION: Our data identify AAA disease-relevant transcriptional signatures of vascular cells in the IAA. Furthermore, we characterize the heterogeneity and cellular responses of VSMCs and monocytes/macrophages during AAA progression, which provide insights into their function and the regulation of AAA onset and progression.

BAF60a Deficiency in Vascular Smooth Muscle Cells Prevents Abdominal Aortic Aneurysm by Reducing Inflammation and Extracellular Matrix Degradation.

OBJECTIVE: Currently, there are no approved drugs for abdominal aortic aneurysm (AAA) treatment, likely due to limited understanding of the primary molecular mechanisms underlying AAA development and progression. BAF60a-a unique subunit of the SWI/SNF (switch/sucrose nonfermentable) chromatin remodeling complex-is a novel regulator of metabolic homeostasis, yet little is known about its function in the vasculature and pathogenesis of AAA. In this study, we sought to investigate the role and underlying mechanisms of vascular smooth muscle cell (VSMC)-specific BAF60a in AAA formation. Approach and Results: BAF60a is upregulated in human and experimental murine AAA lesions. In vivo studies revealed that VSMC-specific knockout of BAF60a protected mice from both Ang II (angiotensin II)-induced and elastase-induced AAA formation with significant suppression of vascular inflammation, monocyte infiltration, and elastin fragmentation. Through RNA sequencing and pathway analysis, we found that the expression of inflammatory response genes in cultured human aortic smooth muscle cells was significantly downregulated by small interfering RNA-mediated BAF60a knockdown while upregulated upon adenovirus-mediated BAF60a overexpression. BAF60a regulates VSMC inflammation by recruiting BRG1 (Brahma-related gene-1)-a catalytic subunit of the SWI/SNF complex-to the promoter region of NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells) target genes. Furthermore, loss of BAF60a in VSMCs prevented the upregulation of the proteolytic enzyme cysteine protease CTSS (cathepsin S), thus ameliorating ECM (extracellular matrix) degradation within the vascular wall in AAA. CONCLUSIONS: Our study demonstrated that BAF60a is required to recruit the SWI/SNF complex to facilitate the epigenetic regulation of VSMC inflammation, which may serve as a potential therapeutic target in preventing and treating AAA.

Multi-omics sequencing provides insight into floral transition in Catalpa bungei. C.A. Mey.

BACKGROUND: Floral transition plays an important role in development, and proper time is necessary to improve the value of valuable ornamental trees. The molecular mechanisms of floral transition remain unknown in perennial woody plants. "Bairihua" is a type of C. bungei that can undergo floral transition in the first planting year. RESULTS: Here, we combined short-read next-generation sequencing (NGS) and single-molecule real-time (SMRT) sequencing to provide a more complete view of transcriptome regulation during floral transition in C. bungei. The circadian rhythm-plant pathway may be the critical pathway during floral transition in early flowering (EF) C. bungei, according to horizontal and vertical analysis in EF and normal flowering (NF) C. bungei. SBP and MIKC-MADS-box were seemingly involved in EF during floral transition. A total of 61 hub genes were associated with floral transition in the MEturquoise model with Weighted Gene Co-expression Network Analysis (WGCNA). The results reveal that ten hub genes had a close connection with the GASA homologue gene (Cbu.gene.18280), and the ten co-expressed genes belong to five flowering-related pathways. Furthermore, our study provides new insights into the complexity and regulation of alternative splicing (AS). The ratio or number of isoforms of some floral transition-related genes is different in different periods or in different sub-genomes. CONCLUSIONS: Our results will be a useful reference for the study of floral transition in other perennial woody plants. Further molecular investigations are needed to verify our sequencing data.