A quick and reliable image-based AI algorithm for evaluating cellular senescence of gastric organoids.

OBJECTIVE: Organoids are a powerful tool with broad application prospects in biomedicine. Notably, they provide alternatives to animal models for testing potential drugs before clinical trials. However, the number of passages for which organoids maintain cellular vitality ex vivo remains unclear. METHODS: Herein, we constructed 55 gastric organoids from 35 individuals, serially passaged the organoids, and captured microscopic images for phenotypic evaluation. Senescence-associated ß-galactosidase (SA-ß-Gal), cell diameter in suspension, and gene expression reflecting cell cycle regulation were examined. The YOLOv3 object detection algorithm integrated with a convolutional block attention module (CBAM) was used to evaluate organoid vitality. RESULTS: SA-ß-Gal staining intensity; single-cell diameter; and expression of p15, p16, p21, CCNA2, CCNE2, and LMNB1 reflected the progression of aging in organoids during passaging. The CBAM-YOLOv3 algorithm precisely evaluated aging organoids on the basis of organoid average diameter, organoid number, and number × diameter, and the findings positively correlated with SA-ß-Gal staining and single-cell diameter. Organoids derived from normal gastric mucosa had limited passaging ability (passages 1-5), before aging, whereas tumor organoids showed unlimited passaging potential for more than 45 passages (511 days) without showing clear senescence. CONCLUSIONS: Given the lack of indicators for evaluating organoid growth status, we established a reliable approach for integrated analysis of phenotypic parameters that uses an artificial intelligence algorithm to indicate organoid vitality. This method enables precise evaluation of organoid status in biomedical studies and monitoring of living biobanks.

Benchmark for establishment of organoids from gastrointestinal epithelium and cancer based on available consumables and reagents.

Gastrointestinal cancers are a public health problem that threatens the lives of human being. A good experimental model is a powerful tool to promote the uncovering pathogenesis and establish novel treatment methods. High-quality biomedical research requires experimental models to recapitulate the physiological and pathological states of their parental tissues as much as possible. Organoids are such experimental models. Organoids refer to small organ-like cellular clusters formed by the expansion and passaging of living tissues in 3D culture medium in vitro. Organoids are highly similar to the original tissues in terms of cellular composition, cell functions, and genomic profiling. Organoids have many advantages, such as short preparation cycles, long-term storage based on cryopreservation, and reusability. In recent years, researchers carried out the establishment of organoids from gastrointestinal mucosa and cancer tissues, and accumulated valuable experiences. In order to promote effective usage and further development of organoid-related technologies in the research of gastrointestinal diseases, this study proposes a benchmark based on utilization of available experimental consumables and reagents, which are involved in the key steps such as collection and pretreatment of biospecimen, organoid construction, organoid cryopreservation and recovery, growth status evaluation, and organoid quality control. We believe that the standard for the construction and preservation of organoids derived from human gastrointestinal epithelium and cancer tissues can provide an important reference for the majority of scientific researchers.

Tracking cancer lesions on surgical samples of gastric cancer by artificial intelligent algorithms.

To quickly locate cancer lesions, especially suspected metastatic lesions after gastrectomy, AI algorithms of object detection and semantic segmentation were established. A total of 509 macroscopic images from 381 patients were collected. The RFB-SSD object detection algorithm and ResNet50-PSPNet semantic segmentation algorithm were used. Another 57 macroscopic images from 48 patients were collected for prospective verification. We used mAP as the metrics of object detection. The best mAP was 95.90% with an average of 89.89% in the test set. The mAP reached 92.60% in validation set. We used mIoU for evaluation of semantic segmentation. The best mIoU was 80.97% with an average of 79.26% in the test set. In addition, 81 out of 92 (88.04%) gastric specimens were accurately predicted for the cancer lesion located at the serosa by ResNet50-PSPNet semantic segmentation model. The positive rate and accuracy of AI prediction were different based on cancer invasive depth. The metastatic lymph nodes were predicted in 24 cases by semantic segmentation model. Among them, 18 cases were confirmed by pathology. The predictive accuracy was 75.00%. Our well-trained AI algorithms effectively identified the subtle features of gastric cancer in resected specimens that may be missed by naked eyes. Taken together, AI algorithms could assist clinical doctors quickly locating cancer lesions and improve their work efficiency.

Dexamethasone suppresses immune evasion by inducing GR/STAT3 mediated downregulation of PD-L1 and IDO1 pathways.

T cell exhaustion plays critical roles in tumor immune evasion. Novel strategies to suppress immune evasion are in urgent need. We aimed to identify potential compounds to target T cell exhaustion and increase response to immune checkpoint inhibitors (ICIs). Differentially expressed genes (DEGs) were identified between tumors with different immune evasion potential by comparing the transcriptome data. DEGs were then analyzed in the Connectivity Map (CMap) platform to identify potential compounds to increase response to ICIs. Gene set enrichment analysis, LDH release assay, Chromatin immunoprecipitation (ChIP), and Co-IP were performed to explore the potential mechanisms in vitro. Patients derived organoids and humanized xenograft mouse model were utilized to validate the finding ex vivo and in vivo. We identified 25 potential compounds that may play critical roles in regulating tumor immune evasion. We further pinpointed a specific compound, dexamethasone, which shows potent anti-tumor effect in multiple cancer cell lines when cocultured with T cells. Dexamethasone can suppress T cell exhaustion by decreasing the activity of two immune checkpoints simultaneously, including PD-L1 and IDO1. Functional study shows dexamethasone can increase the sensitivity of ICIs in coculture system, 3D organoid model and humanized mouse model. Mechanism study shows dexamethasone mediated transcriptional suppression of PD-L1 and IDO1 depends on the nuclear translocation of GR/STAT3 complex. These findings demonstrate dexamethasone can suppress immune evasion by inducing GR/STAT3 mediated downregulation of PD-L1 and IDO1 pathways.

Genetic Profiles Affect the Biological Effects of Serine on Gastric Cancer Cells.

A high serine content in body fluid was identified in a portion of patients with gastric cancer, but its biological significance was not clear. Here, we investigated the biological effect of serine on gastric cancer cells. Serine was added into the culture medium of MGC803 and HGC27 cancer cells, and its influence on multiple biological functions, such as cell growth, migration and invasion, and drug resistance was analyzed. We examined the global transcriptomic profiles in these cultured cells with high serine content. Both MGC803 and HGC27 cell lines were originated from male patients, however, their basal gene expression patterns were very different. The finding of cell differentiation-associated genes, ALPI, KRT18, TM4SF1, KRT81, A2M, MT1E, MUC16, BASP1, TUSC3, and PRSS21 in MGC803 cells suggested that this cell line was more poorly differentiated, compared to HGC27 cell line. When the serine concentration was increased to 150mg/ml in medium, the response of these two gastric cancer cell lines was different, particularly on cell growth, cell migration, and invasion and 5-FU resistance. In animal experiment, administration of high concentration of serine promoted cancer cell metastasis to local lymph node. Taken together, we characterized the basal gene expressing profiles of MGC803 and HGC27. The HGC27 cells were more differentiated than MGC803 cells. MGC803 cells were more sensitive to the change of serine content. Our results suggested that the responsiveness of cancer cells to microenvironmental change is associated with their genetic background.

Glucocorticoids improve severe or critical COVID-19 by activating ACE2 and reducing IL-6 levels.

COVID-19 is a public health emergency that has rapidly spread to over 200 countries and regions, and no effective treatment has been established to date. Severe and critical cases have been associated with higher mortality due to acute respiratory distress syndrome (ARDS) and cytokine storm. Based on the novelty and recent emergence of COVID-19, no effective treatment regimen has been identified, thus prompting clinicians to engage in drug repurposing to address the immediate therapeutic need. This study focused on the molecular target angiotensin-converting enzyme 2 (ACE2) of SARS-CoV-2 and screened a group of ACE2 agonists by bioinformatics. Glucocorticoids are a type of ACE2 activator. We verified the efficacy of nine chemicals on regulating ACE2 expression in human GES-1, an upper digestive tract epithelial cell line, and THP-1, a human monocyte cell line, and found that several glucocorticoids imparted activating effects on ACE2 in both cell lines. The drugs triciribine and kinetin riboside activate ACE2 expression or inhibit IL-6 production in macrophages to some extent. In addition, we compared the efficacies of several glucocorticoids. Hydrocortisone showed the strongest effect on ACE2 activation, followed by prednisolone, dexamethasone, and methylprednisolone. We retrospectively analyzed the therapeutic efficacy of nine severe or critical patients from a cohort of 90 COVID-19 cases, who received medium to small doses of glucocorticoids from our integrated medical team in Wuhan. Seven out of nine patients revealed significant improvement in clinical parameters and chest CT images. This study provides experimental and clinical evidence that medium-to-low-dose glucocorticoids may play a protective role in the respiratory and digestive systems by activating ACE2 and suppressing cytokine storm.

A positive feedback between IDO1 metabolite and COL12A1 via MAPK pathway to promote gastric cancer metastasis.

BACKGROUND: IDO1 (Indoleamine 2,3-dioxygenase 1) inhibits host anti-tumor immune response by exhausting tryptophan in tumor microenvironment, but the pathogenic mechanisms of IDO1 in gastric cancer (GC) cells need to be further explored. METHODS: The aim of this study was to use CCLE (Cancer Cell Line Encyclopedia) transcriptomic data of GC cell lines for WGCNA (Weighted Gene Co-expression Network Analysis) analysis, and explore the potential functions and mechanisms of IDO1 in GC progression in vitro and in vivo. RESULTS: The higher expression level of IDO1 was identified in 4 out of 7 GC cell lines. Increased IDO1 expression strongly promoted cell migration via its metabolite kynurenine and was associated with pathways of immune activation according to GSEA (Gene Set Enrichment Analysis). The functions of IDO1 were closely associated with extracellular matrix, collagen metabolic and catabolic process by WGCNA analysis. Among five hub genes (AXL, SGCE, COL12A1, ANTXR1, LOXL2), COL12A1 and LOXL2 were upregulated in GC tissues. IDO1 disclosed positive correlation with six collagen genes by coefficient matrix diagram. Knockdown of IDO1 decreased the expression of LOXL2, COL6A1, COL6A2 and COL12A1 in GC cells in both mRNA and protein levels. Of them, knockdown of COL12A1 inhibited cell migration more apparently than knockdown of others. IDO1 and COL12A1 revealed synergistic efficacy on promoting cell migration via a positive feedback sustained by MAPK pathway. This bioprocess was mediated by IDO1 metabolite kynurenine and integrin ß1. A popliteal lymph nodemetastasis model was established for verifying metastatic promotion of IDO1 and COL12A1 in GC. CONCLUSIONS: IDO1 and COL12A1 synergistically promoted GC metastasis. The novel findings suggested that both IDO1 and COL12A1 may be promising targets on anti-cancer treatment in GC.

A Novel Citrullinated Modification of Histone 3 and Its Regulatory Mechanisms Related to IPO-38 Antibody-Labeled Protein.

IPO-38 is a potential biomarker for early diagnosis of gastric cancer that we recently identified. Although we characterized its chemical nature as a nucleosome histone, we suspected the existence of histone modification for the IPO-38 antibody-labeled protein. Here, we used a commercially available modified histone peptide array to identify the type and site of histone modification labeled by the IPO-38 monoclonal antibody. In protein array analysis, the citrulline modification of histone 3 on arginine 26 (H3R26Cit) yielded the strongest signal. Although peptidyl arginine deiminase-2 and -4 (PADI2 and PADI4, respectively) can catalyze the conversion of arginine to citrulline, we observed that only PADI4 expression correlated with the citrulline histone modification of H3R26Cit. Overexpression of PADI4, via transfection of a eukaryotic expression vector, and knockdown of PADI4 gene expression, by a PADI4 CRISPR/Cas9 vector, confirmed the crucial function of PADI4 on the increased level of H3R26Cit in gastric cancer cell lines. By immunoprecipitation and immunoblotting, we found an interaction between H3R26Cit and H3K27me3. Our study established the first link between the IPO-38 antigen and citrullinated histone 3, and clarified the upstream regulatory enzyme PADI4. The new findings suggest an important role for the citrullination modification of histone in gastric cancer biology, and should help us optimize the development of a sensitive and specific diagnostic reagent.

Cross-Database Analysis Reveals Sensitive Biomarkers for Combined Therapy for ERBB2+ Gastric Cancer.

Exploring ERBB2-related pathways will help us finding sensitive molecules and potential combined therapeutic targets of ERBB2-targeted therapy for ERBB2+ gastric cancer (GC). In this study, we performed a cross-databases study focused on ERBB2+ GC. The data of ERBB2+ GC deposited in the cancer genome atlas (TCGA), gene expression omnibus (GEO), InBio MapTM, cancer cell line encyclopedia (CCLE), and cancer therapeutics response portal (CTRP) were analyzed. The correlation of expression levels of candidate and IC50 of candidate genes-targeted drugs were verified on NCI-N87 and MKN-45 GC cell lines. We found that RARA, THRA, CACNB1, and TOP2A are drug sensitive biomarkers of ERBB2-targeted treatment with FDA-approved drugs. All these genes act through Myc signaling pathway. Myc is the downstream hub gene of both ERBB2 and RARA. The expression of RARA, THRA, and CACNB1 were negatively correlated with Myc activation, while ERBB2 and TOP2A positively correlated with Myc activation. SH3BGRL3, SH3BGRL, and NRG2 were identified as potential ligands of ERBB2. The ERBB2+ GC with RARA amplification demonstrated better prognosis than those without RARA amplification, while overexpression of NRG2 and SH3BGRL correlated with poor prognosis in ERBB2+ GC. About 90% of ERBB2+ GC was compatible with chromosome instability (CIN) subtype of TCGA, which overlaps with intestinal-type GC in Lauren classification. In validating experiments, combination of Lapatinib and all-trans retinoic acid (ATRA) synergistically suppresses cell growth, and accompanied by decreased expression of MYC. In conclusions, we identified several predicting biomarkers for ERBB2-targeted therapy and corresponding histological features of ERBB2+ GC. Combination of ERBB2 antagonist or RARA agonist may be effective synergistic regimens for ERBB2+ GC.

Protein arginine methyltransferase 5-mediated epigenetic silencing of IRX1 contributes to tumorigenicity and metastasis of gastric cancer.

IRX1 is originally characterized as a tumor suppressor gene of gastric cancer (GC) by our group based on serially original studies. However, the molecular regulatory mechanisms of IRX1 are not clear yet. Here, we identified protein arginine methyltransferase 5 (PRMT5) as a major upstream regulator of IRX1 for determining GC progression. Expression of PRMT5 was significantly increased in human GC tissues (433 out of 602 cases, 71.93%) compared with normal gastric mucosa, and exhibited diagnostic and prognostic potential. Overexpression of PRMT5 promoted tumorigenicity and metastasis of GC cells, while knockdown of PRMT5 abrogated tumorigenicity and metastasis of GC cells in vitro and in vivo. By co-immunoprecipitation and chromatin immunoprecipitation assays, we proved that PRMT5 elevated methylation levels of tumor suppressor IRX1 promoter via recruiting DNMT3A at promoter region. Knockdown of PRMT5 in SGC7901 and NCI-N87 cells decreased the recruitment of DNMT3A at IRX1 promoter, and reduced the methylation level of IRX1 promoter, then re-activated IRX1 expression. Whereas, overexpression of PRMT5 could epigenetically suppress IRX1 expression. Overall, PRMT5 promoted tumorigenicity and metastasis of gastric cancer cells via epigenetic silencing of IRX1. Targeting PRMT5 in GC might inhibit the malignant characters of GC and drawing a novel therapeutic potential.