Liver fibrosis is a very common condition seen in millions of patients with various liver diseases, and yet no effective treatments are available owing to poorly characterized molecular pathogenesis. Here, we show that leukocyte cell-derived chemotaxin 2 (LECT2) is a functional ligand of Tie1, a poorly characterized endothelial cell (EC)-specific orphan receptor. Upon binding to Tie1, LECT2 interrupts Tie1/Tie2 heterodimerization, facilitates Tie2/Tie2 homodimerization, activates PPAR signaling, and inhibits the migration and tube formations of EC. In vivo studies showed that LECT2 overexpression inhibits portal angiogenesis, promotes sinusoid capillarization, and worsens fibrosis, whereas these changes were reversed in Lect2-KO mice. Adeno-associated viral vector serotype 9 (AAV9)-LECT2 small hairpin RNA (shRNA) treatment significantly attenuates fibrosis. Upregulation of LECT2 is associated with advanced human liver fibrosis staging. We concluded that targeting LECT2/Tie1 signaling may represent a potential therapeutic target for liver fibrosis, and serum LECT2 level may be a potential biomarker for the screening and diagnosis of liver fibrosis. (A) LECT2 inhibited the migration of EA.hy926 cell. Scale bar, 100 mm. (B) LECT2 inhibited tube formation of EA.hy926 cell. Scale bar, 200 mm. (C) Inhibition of LECT2 enhanced migration of EA.hy926 cell. Scale bar, 100 mm. (D) Inhibition of LECT2 enhanced tube formation of EA.hy926 cell. Scale bar, 200 mm. (E) LECT2 inhibited the migration of primary liver sinusoid endothelial cell (LSEC). Scale bar, 100 mm. (F) LECT2 inhibited tube formation of primary LSEC. Scale bar, 200 mm. (G) Liver tissues were immunohistochemically stained for CD31. Arrows indicate portal vessels, and arrowheads indicate capillarization of liver sinusoids. Scale bar, 200 mm. (H) The number of CD31-positive vessels surrounding the portal area was measured. (I) The CD31-positive capillarization of liver sinusoids was measured. (J) Liver tissues were immunohistochemically stained for CD31. Arrows indicate portal vessels, and arrowheads indicate capillarization of liver sinusoids. Scale bar, 200 mm. (K) The number of CD31-positive vessels surrounding the portal area was measured. (L) The CD31-positive capillarization of liver sinusoids was measured.

Induction of endogenous adult stem cells by administering soluble molecules provides an advantageous approach for tissue damage repair, which could be a clinically applicable and cost-effective alternative to transplantation of embryonic or pluripotent stem cell-derived tissues for the treatment of acute organ failures. Here, we show that HGF/Rspo1 induce liver stem cells and rescue liver dysfunction. Carbon tetrachloride treatment promotes both fibrosis and Lgr5+ liver stem cell proliferation, whereas Lgr5 knockdown worsens fibrosis. Injection of HGF in combination with Rspo1 increases the number of Lgr5+ liver stem cells and improves liver function by attenuating fibrosis. We observe Lgr5+ liver stem cells in human liver fibrosis tissues, and once they are isolated, these cells are able to form organoids, and treatment with HGF/Rspo1 promotes their expansion. We suggest that Lgr5+ liver stem cells represent a valuable target for liver damage treatment, and that HGF/Rspo1 can be used to promote liver stem cell expansion.

The silver surface plasmon resonance (SPR) sensor has long been explored due to its intrinsic sensitivity enhancement over the conventional single-layered gold SPR sensor. However, the silver SPR sensor has not been exploited for practical applications because of pronounced instability problems. We propose a novel gold-silver-gold trilayered SPR sensor chip, in which an extra buffer layer of gold is added between the silver and substrate adhesion layer (i.e., chromium) compared to the previously reported silver-gold bilayered SPR sensors. Subjected to prolonged agitation in phosphate-buffered saline (PBS) solution, the new chip exhibited high integrity according to both optical and atomic force microscopy (AFM) analysis. Having undergone repeated cycles of calibration, binding, and regeneration in various chemical solutions, 25 regions of interest (ROIs) over a 14 mm ×14 mm area were chosen and monitored by large detection area SPR microscopy; the new sensor chip exhibited stability comparable to the single gold layered SPR chip. In terms of sensing performances, over 50% increases in sensitivity and signal-to-noise ratio (S/N) than those of the single gold layered SPR chip were determined by SPR microscopy at 660 nm. Protein arrays of protein A and bovine serum albumin (BSA) on both the new chip and single-layered gold SPR chip were fabricated and underwent biomolecular interactions with human IgG, for the purpose of consistency, comparison on kinetics parameters, values from the microarray trilayered chip showed reasonable consistency with those from the single gold layered SPR chip. This study suggests that the new chip is a viable alternative to the conventional single gold layered SPR chip with improved sensing performances.

Background Targeted therapy and immunotherapy put forward higher demands for accurate lung cancer classification, as well as benign versus malignant disease discrimination. Digital whole slide images (WSIs) witnessed the transition from traditional histopathology to computational approaches, arousing a hype of deep learning methods for histopathological analysis. We aimed at exploring the potential of deep learning models in the identification of lung cancer subtypes and cancer mimics from WSIs. Methods We initially obtained 741 WSIs from the First Affiliated Hospital of Sun Yat-sen University (SYSUFH) for the deep learning model development, optimization, and verification. Additional 318 WSIs from SYSUFH, 212 from Shenzhen People’s Hospital, and 422 from The Cancer Genome Atlas were further collected for multi-centre verification. EfficientNet-B5- and ResNet-50-based deep learning methods were developed and compared using the metrics of recall, precision, F1-score, and areas under the curve (AUCs). A threshold-based tumour-first aggregation approach was proposed and implemented for the label inferencing of WSIs with complex tissue components. Four pathologists of different levels from SYSUFH reviewed all the testing slides blindly, and the diagnosing results were used for quantitative comparisons with the best performing deep learning model. Results We developed the first deep learning-based six-type classifier for histopathological WSI classification of lung adenocarcinoma, lung squamous cell carcinoma, small cell lung carcinoma, pulmonary tuberculosis, organizing pneumonia, and normal lung. The EfficientNet-B5-based model outperformed ResNet-50 and was selected as the backbone in the classifier. Tested on 1067 slides from four cohorts of different medical centres, AUCs of 0.970, 0.918, 0.963, and 0.978 were achieved, respectively. The classifier achieved high consistence to the ground truth and attending pathologists with high intraclass correlation coefficients over 0.873. Conclusions Multi-cohort testing demonstrated our six-type classifier achieved consistent and comparable performance to experienced pathologists and gained advantages over other existing computational methods. The visualization of prediction heatmap improved the model interpretability intuitively. The classifier with the threshold-based tumour-first label inferencing method exhibited excellent accuracy and feasibility in classifying lung cancers and confused nonneoplastic tissues, indicating that deep learning can resolve complex multi-class tissue classification that conforms to real-world histopathological scenarios.