Zika Virus Causes Testis Damage and Leads to Male Infertility in Mice.

Zika virus (ZIKV) persists in the semen of male patients, a first for flavivirus infection. Here, we demonstrate that ZIKV can induce inflammation in the testis and epididymidis, but not in the prostate or seminal vesicle, and can lead to damaged testes after 60 days post-infection in mice. ZIKV induces innate immune responses in Leydig, Sertoli, and epididymal epithelial cells, resulting in the production of pro-inflammatory cytokines/chemokines. However, ZIKV does not induce a rapid and abundant cytokine production in peritubular cell and spermatogonia, suggesting that these cells are vulnerable for ZIKV infection and could be the potential repositories for ZIKV. Our study demonstrates a correlation between ZIKV and testis infection/damage and suggests that ZIKV infection, under certain circumstances, can eventually lead to male infertility.

Osmosensor-mediated control of Ca2+ spiking in pollen germination.

Higher plants survive terrestrial water deficiency and fluctuation by arresting cellular activities (dehydration) and resuscitating processes (rehydration). However, how plants monitor water availability during rehydration is unknown. Although increases in hypo-osmolarity-induced cytosolic Ca2+ concentration (HOSCA) have long been postulated to be the mechanism for sensing hypo-osmolarity in rehydration1,2, the molecular basis remains unknown. Because osmolarity triggers membrane tension and the osmosensing specificity of osmosensing channels can only be determined in vivo3-5, these channels have been classified as a subtype of mechanosensors. Here we identify bona fide cell surface hypo-osmosensors in Arabidopsis and find that pollen Ca2+ spiking is controlled directly by water through these hypo-osmosensors-that is, Ca2+ spiking is the second messenger for water status. We developed a functional expression screen in Escherichia coli for hypo-osmosensitive channels and identified OSCA2.1, a member of the hyperosmolarity-gated calcium-permeable channel (OSCA) family of proteins6. We screened single and high-order OSCA mutants, and observed that the osca2.1/osca2.2 double-knockout mutant was impaired in pollen germination and HOSCA. OSCA2.1 and OSCA2.2 function as hypo-osmosensitive Ca2+-permeable channels in planta and in HEK293 cells. Decreasing osmolarity of the medium enhanced pollen Ca2+ oscillations, which were mediated by OSCA2.1 and OSCA2.2 and required for germination. OSCA2.1 and OSCA2.2 convert extracellular water status into Ca2+ spiking in pollen and may serve as essential hypo-osmosensors for tracking rehydration in plants.

Human pluripotent stem cell-derived inner ear organoids recapitulate otic development in vitro.

Our molecular understanding of the early stages of human inner ear development has been limited by the difficulty in accessing fetal samples at early gestational stages. As an alternative, previous studies have shown that inner ear morphogenesis can be partially recapitulated using induced pluripotent stem cells directed to differentiate into inner ear organoids (IEOs). Once validated and benchmarked, these systems could represent unique tools to complement and refine our understanding of human otic differentiation and model developmental defects. Here, we provide the first direct comparisons of the early human embryonic otocyst and fetal sensory organs with human IEOs. We use multiplexed immunostaining and single-cell RNA-sequencing to characterize IEOs at three key developmental steps, providing a new and unique signature of in vitro-derived otic placode, epithelium, neuroblasts and sensory epithelia. In parallel, we evaluate the expression and localization of crucial markers at these equivalent stages in human embryos. Together, our data indicate that the current state-of-the-art protocol enables the specification of bona fide otic tissue, supporting the further application of IEOs to inform inner ear biology and disease.

Mapping oto-pharyngeal development in a human inner ear organoid model.

Inner ear development requires the coordination of cell types from distinct epithelial, mesenchymal and neuronal lineages. Although we have learned much from animal models, many details about human inner ear development remain elusive. We recently developed an in vitro model of human inner ear organogenesis using pluripotent stem cells in a 3D culture, fostering the growth of a sensorineural circuit, including hair cells and neurons. Despite previously characterizing some cell types, many remain undefined. This study aimed to chart the in vitro development timeline of the inner ear organoid to understand the mechanisms at play. Using single-cell RNA sequencing at ten stages during the first 36 days of differentiation, we tracked the evolution from pluripotency to various ear cell types after exposure to specific signaling modulators. Our findings showcase gene expression that influences differentiation, identifying a plethora of ectodermal and mesenchymal cell types. We also discern aspects of the organoid model consistent with in vivo development, while highlighting potential discrepancies. Our study establishes the Inner Ear Organoid Developmental Atlas (IODA), offering deeper insights into human biology and improving inner ear tissue differentiation.

SIRT2 promotes base excision repair by transcriptionally activating OGG1 in an ATM/ATR-dependent manner.

Sirtuin 2 (SIRT2) regulates the maintenance of genome integrity by targeting pathways of DNA damage response and homologous recombination repair. However, whether and how SIRT2 promotes base excision repair (BER) remain to be determined. Here, we found that independent of its catalytic activity SIRT2 interacted with the critical glycosylase OGG1 to promote OGG1 recruitment to its own promoter upon oxidative stress, thereby enhancing OGG1 promoter activity and increasing BER efficiency. Further studies revealed that SIRT2 was phosphorylated on S46 and S53 by ATM/ATR upon oxidative stress, and SIRT2 phosphorylation enhanced the SIRT2-OGG1 interaction and mediated the stimulatory effect of SIRT2 on OGG1 promoter activity. We also characterized 37 cancer-derived SIRT2 mutants and found that 5 exhibited the loss of the stimulatory effects on OGG1 transcription. Together, our data reveal that SIRT2 acts as a tumor suppressor by promoting OGG1 transcription and increasing BER efficiency in an ATM/ATR-dependent manner.

Thrombocytopenia in SFTS due to platelets with altered function undergoing cell death pathways.

BACKGROUND: Thrombocytopenia is the major clinical feature associated with the severity of SFTS, but the mechanism by which it occurs remains unclear. METHODS: RNA transcriptome analyses were performed on platelets purified from SFTS patients and SFTSV-infected mice. The functions of differentially expressed genes (DEGs) in the platelets were characterized. ELISA, flow cytometry, and qRT-PCR were used to measure the levels of platelet activation, SFTSV infection in platelets, formation of neutrophil extracellular traps (NETs), transcription of DEGs and percent of platelets undergoing cell death. RESULTS: Enhanced neutrophil activation and interferon (IFN) signaling involved in the viral life cycle were common platelet responses in SFTS, which may consume increasing numbers of platelets. Other functional changes may be associated with different outcomes of SFTS. SFTSV infection led to platelet destruction by pyroptosis, apoptosis, necroptosis, and autophagy. In contrast to SFTS patients, platelets in SFTSV-infected mice mainly play a role in adaptive immunity, and platelet death was not as severe as in humans. CONCLUSIONS: The altered functions of platelets, such as mediating leukocyte activation and undergoing cell death, contribute to thrombocytopenia in SFTS patients. The different mechanisms of thrombocytopenia in mice, suggest that platelet functions should be considered in experimental animal models.

YTHDF1 facilitates PRC1-mediated H2AK119ub in human ES cells.

Polycomb repressive complexes (PRCs) play critical roles in cell fate decisions during normal development as well as disease progression through mediating histone modifications such as H3K27me3 and H2AK119ub. How exactly PRCs recruited to chromatin remains to be fully illuminated. Here, we report that YTHDF1, the N6-methyladenine (m6 A) RNA reader that was previously known to be mainly cytoplasmic, associates with RNF2, a PRC1 protein that mediates H2AK119ub in human embryonic stem cells (hESCs). A portion of YTHDF1 localizes in the nuclei and associates with RNF2/H2AK119ub on a subset of gene loci related to neural development functions. Knock-down YTHDF1 attenuates H2AK119ub modification on these genes and promotes neural differentiation in hESCs. Our findings provide a noncanonical mechanism that YTHDF1 participates in PRC1 functions in hESCs.

Single-cell proteomics differentiates Arabidopsis root cell types.

Single-cell proteomics (SCP) is an emerging approach to resolve cellular heterogeneity within complex tissues of multi-cellular organisms. Here, we demonstrate the feasibility of SCP on plant samples using the model plant Arabidopsis thaliana. Specifically, we focused on examining isolated single cells from the cortex and endodermis, which are two adjacent root cell types derived from a common stem cell lineage. From 756 root cells, we identified 3763 proteins and 1118 proteins/cell. Ultimately, we focus on 3217 proteins quantified following stringent filtering. Of these, we identified 596 proteins whose expression is enriched in either the cortex or endodermis and are able to differentiate these closely related plant cell types. Collectivity, this study demonstrates that SCP can resolve neighboring cell types with distinct functions, thereby facilitating the identification of biomarkers and candidate proteins to enable functional genomics.

CACNA1B protects naked mole-rat hippocampal neuron from apoptosis via altering the subcellular localization of Nrf2 after 60Co irradiation.

Although radiotherapy is the most effective treatment modality for brain tumors, it always injures the central nervous system, leading to potential sequelae such as cognitive dysfunction. Radiation induces molecular, cellular, and functional changes in neuronal and glial cells. The hippocampus plays a critical role in learning and memory; therefore, concerns about radiation-induced injury are widespread. Multiple studies have focused on this complex problem, but the results have not been fully elucidated. Naked mole rat brains were irradiated with 60Co at a dose of 10 Gy. On 7 days, 14 days, and 28 days after irradiation, hippocampi in the control groups were obtained for next-generation sequencing. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were subsequently performed. Venn diagrams revealed 580 differentially expressed genes (DEGs) that were common at different times after irradiation. GO and KEGG analyses revealed that the 580 common DEGs were enriched in molecular transducer activity. In particular, CACNA1B mediated regulatory effects after irradiation. CACNA1B expression increased significantly after irradiation. Downregulation of CACNA1B led to a reduction in apoptosis and reactive oxygen species levels in hippocampal neurons. This was due to the interaction between CACNA1B and Nrf2, which disturbed the normal nuclear localization of Nrf2. In addition, CACNA1B downregulation led to a decrease in the cognitive functions of naked mole rats. These findings reveal the pivotal role of CACNA1B in regulating radiation-induced brain injury and will lead to the development of a novel strategy to prevent brain injury after irradiation.

Autophagy-dependent lysosomal calcium overload and the ATP5B-regulated lysosomes-mitochondria calcium transmission induce liver insulin resistance under perfluorooctane sulfonate exposure.

Perfluorooctane sulfonate (PFOS), an officially listed persistent organic pollutant, is a widely distributed perfluoroalkyl substance. Epidemiological studies have shown that PFOS is intimately linked to the occurrence of insulin resistance (IR). However, the detailed mechanism remains obscure. In previous studies, we found that mitochondrial calcium overload was concerned with hepatic IR induced by PFOS. In this study, we found that PFOS exposure noticeably raised lysosomal calcium in L-02 hepatocytes from 0.5 h. In the PFOS-cultured L-02 cells, inhibiting autophagy alleviated lysosomal calcium overload. Inhibition of mitochondrial calcium uptake aggravated the accumulation of lysosomal calcium, while inhibition of lysosomal calcium outflowing reversed PFOS-induced mitochondrial calcium overload and IR. Transient receptor potential mucolipin 1 (TRPML1), the calcium output channel of lysosomes, interacted with voltage-dependent anion channel 1 (VDAC1), the calcium intake channel of mitochondria, in the PFOS-cultured cells. Moreover, we found that ATP synthase F1 subunit beta (ATP5B) interacted with TRPML1 and VDAC1 in the L-02 cells and the liver of mice under PFOS exposure. Inhibiting ATP5B expression or restraining the ATP5B on the plasma membrane reduced the interplay between TRPML1 and VDAC1, reversed the mitochondrial calcium overload and deteriorated the lysosomal calcium accumulation in the PFOS-cultured cells. Our research unveils the molecular regulation of the calcium crosstalk between lysosomes and mitochondria, and explains PFOS-induced IR in the context of activated autophagy.

Sodium arsenite induces hepatic stellate cells activation by m6A modification of TGF-ß1 during liver fibrosis.

The compound known as Sodium arsenite (NaAsO2), which is a prevalent type of inorganic arsenic found in the environment, has been strongly associated with liver fibrosis (LF), a key characteristic of nonalcoholic fatty liver disease (NAFLD), which has been demonstrated in our previous study. Our previous research has shown that exposure to NaAsO2 triggers the activation of hepatic stellate cells (HSCs), a crucial event in the development of LF. However, the molecular mechanism is still unknown. N6-methyladenosine (m6A) modification is the most crucial post-transcriptional modification in liver disease. Nevertheless, the precise function of m6A alteration in triggering HSCs and initiating LF caused by NaAsO2 remains unknown. Here, we found that NaAsO2 induced LF and HSCs activation through TGF-ß/Smad signaling, which could be reversed by TGF-ß1 knockdown. Furthermore, NaAsO2 treatment enhanced the m6A modification level both in vivo and in vitro. Significantly, NaAsO2 promoted the specific interaction of METTL14 and IGF2BP2 with TGF-ß1 and enhanced the TGF-ß1 mRNA stability. Notably, NaAsO2-induced TGF-ß/Smad pathway and HSC-t6 cells activation might be avoided by limiting METTL14/IGF2BP2-mediated m6A modification. Our findings showed that the NaAsO2-induced activation of HSCs and LF is made possible by the METTL14/IGF2BP2-mediated m6A methylation of TGF-ß1, which may open up new therapeutic options for LF brought on by environmental hazards.

The VDAC1 oligomerization regulated by ATP5B leads to the NLRP3 inflammasome activation in the liver cells under PFOS exposure.

As a persistent organic pollutant, perfluorooctane sulfonate (PFOS) has a serious detrimental impact on human health. It has been suggested that PFOS is associated with liver inflammation. However, the underlying mechanisms are still unclear. Here, PFOS was found to elevate the oligomerization tendency of voltage-dependent anion channel 1 (VDAC1) in the mice liver and human normal liver cells L-02. Inhibition of VDAC1 oligomerization alleviated PFOS-induced nucleotide-binding domain and leucine-rich repeat protein-3 (NLRP3) inflammasome activation. Cytoplasmic membrane VDAC1 translocated to mitochondria was also observed in response to PFOS. Therefore, the oligomerization of VDAC1 occurred mainly in the mitochondria. VDAC1 was found to interact with the ATP synthase beta subunit (ATP5B) under PFOS treatment. Knockdown of ATP5B or immobilization of ATP5B to the cytoplasmic membrane alleviated the increased VDAC1 oligomerization and NLRP3 inflammasome activation. Therefore, our results suggested that PFOS induced NLRP3 inflammasome activation through VDAC1 oligomerization, a process dependent on ATP5B to transfer VDAC1 from the plasma membrane to the mitochondria. The findings offer novel perspectives on the activation of the NLRP3 inflammasome, the regulatory mode on VDAC1 oligomerization, and the mechanism of PFOS toxicity.

Genomic analyses provide insights into the polyploidization-driven herbicide adaptation in Leptochloa weeds.

Polyploidy confers a selective advantage under stress conditions; however, whether polyploidization mediates enhanced herbicide adaptation remains largely unknown. Tetraploid Leptochloa chinensis is a notorious weed in the rice ecosystem, causing severe yield loss in rice. In China, L. chinensis has only one sister species, the diploid L. panicea, whose damage is rarely reported. To gain insights into the effects of polyploidization on herbicide adaptation, we first assembled a high-quality genome of L. panicea and identified genome structure variations with L. chinensis. Moreover, we identified herbicide-resistance genes specifically expanded in L. chinensis, which may confer a greater herbicide adaptability in L. chinensis. Analysis of gene retention and loss showed that five herbicide target-site genes and several herbicide nontarget-site resistance gene families were retained during polyploidization. Notably, we identified three pairs of polyploidization-retained genes including LcABCC8, LcCYP76C1 and LcCYP76C4 that may enhance herbicide resistance. More importantly, we found that both copies of LcCYP76C4 were under herbicide selection during the spread of L. chinensis in China. Furthermore, we identified another gene potentially involved in herbicide resistance, LcCYP709B2, which is also retained during polyploidization and under selection. This study provides insights into the genomic basis of the enhanced herbicide adaptability of Leptochloa weeds during polyploidization and provides guidance for the precise and efficient control of polyploidy weeds.

Single image relighting based on illumination field reconstruction.

Relighting a single low-light image is a crucial and challenging task. Previous works primarily focused on brightness enhancement but neglected the differences in light and shadow variations, which leads to unsatisfactory results. Herein, an illumination field reconstruction (IFR) algorithm is proposed to address this issue by leveraging physical mechanism guidance, physical-based supervision, and data-based modeling. Firstly, we derived the Illumination field modulation equation as a physical prior to guide the network design. Next, we constructed a physical-based dataset consisting of image sequences with diverse illumination levels as supervision. Finally, we proposed the IFR neural network (IFRNet) to model the relighting progress and reconstruct photorealistic images. Extensive experiments demonstrate the effectiveness of our method on both simulated and real-world datasets, showing its generalization ability in real-world scenarios, even training solely from simulation.

AS3MT facilitates NLRP3 inflammasome activation by m6A modification during arsenic-induced hepatic insulin resistance.

N6-methyladenosine (m6A) messenger RNA methylation is the most widespread gene regulatory mechanism affecting liver functions and disorders. However, the relationship between m6A methylation and arsenic-induced hepatic insulin resistance (IR), which is a critical initiating event in arsenic-induced metabolic syndromes such as type 2 diabetes (T2D) and non-alcoholic fatty liver disease (NAFLD), remains unclear. Here, we showed that arsenic treatment facilitated methyltransferase-like 14 (METTL14)-mediated m6A methylation, and that METTL14 interference reversed arsenic-impaired hepatic insulin sensitivity. We previously showed that arsenic-induced NOD-like receptor protein 3 (NLRP3) inflammasome activation contributed to hepatic IR. However, the regulatory mechanisms underlying the role of arsenic toward the post-transcriptional modification of NLRP3 remain unclear. Here, we showed that NLRP3 mRNA stability was enhanced by METTL14-mediated m6A methylation during arsenic-induced hepatic IR. Furthermore, we demonstrated that arsenite methyltransferase (AS3MT), an essential enzyme in arsenic metabolic processes, interacted with NLRP3 to activate the inflammasome, thereby contributing to arsenic-induced hepatic IR. Also, AS3MT strengthened the m6A methylase association with NLRP3 to stabilize m6A-modified NLRP3. In summary, we showed that AS3MT-induced m6A modification critically regulated NLRP3 inflammasome activation during arsenic-induced hepatic IR, and we identified a novel post-transcriptional function of AS3MT in promoting arsenicosis.

Network Analysis Reveals a Distinct Axis of Macrophage Activation in Response to Conflicting Inflammatory Cues.

Macrophages are subject to a wide range of cytokine and pathogen signals in vivo, which contribute to differential activation and modulation of inflammation. Understanding the response to multiple, often-conflicting cues that macrophages experience requires a network perspective. In this study, we integrate data from literature curation and mRNA expression profiles obtained from wild type C57/BL6J mice macrophages to develop a large-scale computational model of the macrophage signaling network. In response to stimulation across all pairs of nine cytokine inputs, the model predicted activation along the classic M1-M2 polarization axis but also a second axis of macrophage activation that distinguishes unstimulated macrophages from a mixed phenotype induced by conflicting cues. Along this second axis, combinations of conflicting stimuli, IL-4 with LPS, IFN-γ, IFN-ß, or TNF-α, produced mutual inhibition of several signaling pathways, e.g., NF-κB and STAT6, but also mutual activation of the PI3K signaling module. In response to combined IFN-γ and IL-4, the model predicted genes whose expression was mutually inhibited, e.g., iNOS or Nos2 and Arg1, or mutually enhanced, e.g., Il4rα and Socs1, validated by independent experimental data. Knockdown simulations further predicted network mechanisms underlying functional cross-talk, such as mutual STAT3/STAT6-mediated enhancement of Il4rα expression. In summary, the computational model predicts that network cross-talk mediates a broadened spectrum of macrophage activation in response to mixed pro- and anti-inflammatory cytokine cues, making it useful for modeling in vivo scenarios.

Accuracy of intravascular ultrasound-derived virtual fractional flow reserve (FFR) and FFR derived from computed tomography for functional assessment of coronary artery disease.

BACKGROUND: Coronary computed tomography-derived fractional flow reserve (CT-FFR) and intravascular ultrasound-derived fractional flow reserve (IVUS-FFR) are two functional assessment methods for coronary stenoses. However, the calculation algorithms for these methods differ significantly. This study aimed to compare the diagnostic performance of CT-FFR and IVUS-FFR using invasive fractional flow reserve (FFR) as the reference standard. METHODS: Six hundred and seventy patients (698 lesions) with known or suspected coronary artery disease were screened for this retrospective analysis between January 2020 and July 2021. A total of 40 patients (41 lesions) underwent intravascular ultrasound (IVUS) and FFR evaluations within six months after completing coronary CT angiography were included. Two novel CFD-based models (AccuFFRct and AccuFFRivus) were used to compute the CT-FFR and IVUS-FFR values, respectively. The invasive FFR ≤ 0.80 was used as the reference standard for evaluating the diagnostic performance of CT-FFR and IVUS-FFR. RESULTS: Both AccuFFRivus and AccuFFRct demonstrated a strong correlation with invasive FFR (R = 0.7913, P < 0.0001; and R = 0.6296, P < 0.0001), and both methods showed good agreement with FFR. The area under the receiver operating characteristic curve was 0.960 (P < 0.001) for AccuFFRivus and 0.897 (P < 0.001) for AccuFFRct in predicting FFR ≤ 0.80. FFR ≤ 0.80 were predicted with high sensitivity (96.6%), specificity (85.7%), and the Youden index (0.823) using the same cutoff value of 0.80 for AccuFFRivus. A good diagnostic performance (sensitivity 89.7%, specificity 85.7%, and Youden index 0.754) was also demonstrated by AccuFFRct. CONCLUSIONS: AccuFFRivus, computed from IVUS images, exhibited a high diagnostic performance for detecting myocardial ischemia. It demonstrated better diagnostic power than AccuFFRct, and could serve as an accurate computational tool for ischemia diagnosis and assist in clinical decision-making.

The effects of free Cys residues on the structure, activity, and tetrameric stability of mammalian uricase.

Mammalian uricases contain four conserved cysteine (Cys) residues, but little is known about their structures and functions. In this study, we first confirmed that all four Cys residues are free and not involved in disulfide bond formation, using canine uricase as a model protein. Cys residues had a greater effect on stability than on activity based on single Cys-to-Ser (serine) substitutions. Circular dichroism (CD) and homology modeling indicated that C188S reduces ß-sheet contents and inter- and intra-subunit hydrophobic interaction, potentially impairing the core tetrameric ß-barrel structure of the tunneling-fold protein, and ultimately decreased the tetrameric stability. Additionally, the inactivation of C188S during the stability tests may be a complex process involving depolymerization followed by irregular aggregation. Double mutations or thiol blockage of Cys188 and Cys195 significantly disrupted the formation and stability of tetrameric uricase, which may be mediated by the free thiols in Cys residues. The present results demonstrated that the free Cys residues are essential for tetrameric formation and stability in mammalian uricase. This implies that free cysteine residues, although not involved in disulfide bonding, may play important structural roles in certain proteins, underscoring the significance of the hydrophobic characteristics of the free thiols in Cys residues. KEY POINTS: • Four Cys residues are not involved in disulfide bonding in mammalian uricase. • The hydrophobicity of free thiols is critical for tetrameric stability in uricase. • Free Cys residues can serve structural roles without involving in disulfide bonds.

Deep learning-based multi-model approach on electron microscopy image of renal biopsy classification.

BACKGROUND: Electron microscopy is important in the diagnosis of renal disease. For immune-mediated renal disease diagnosis, whether the electron-dense granule is present in the electron microscope image is of vital importance. Deep learning methods perform well at feature extraction and assessment of histologic images. However, few studies on deep learning methods for electron microscopy images of renal biopsy have been published. This study aimed to develop a deep learning-based multi-model to automatically detect whether the electron-dense granule is present in the TEM image of renal biopsy, and then help diagnose immune-mediated renal disease. METHODS: Three deep learning models are trained to classify whether the electron-dense granule is present using 910 electron microscopy images of renal biopsies. We proposed two novel methods to improve the model accuracy. One model uses the pre-trained ResNet convolutional layers for feature extraction with transfer learning which was firstly improved with skip architecture, then uses Support Vector Machine as the classifier. We developed a multi-model to combine the traditional ResNet model with the improved one to further improve the accuracy. RESULTS: Deep learning-based multi-model has the highest model accuracy, and the average accuracy is about 88%. The improved ReseNet + SVM model performance is much better than the traditional ResNet model. The average accuracy of the improved ResNet + SVM model is 83%, while the traditional ResNet model accuracy is only 58%. CONCLUSIONS: This study presents the first models for electron microscopy image classification of Renal Biopsy. Identifying whether the electron-dense granule is present plays an important role in the diagnosis of immune complex nephropathy. This study made it possible for Artificial Intelligence models assist to analyze complex electron microscopy images for disease diagnosis.