Early human lung immune cell development and its role in epithelial cell fate.

Studies of human lung development have focused on epithelial and mesenchymal cell types and function, but much less is known about the developing lung immune cells, even though the airways are a major site of mucosal immunity after birth. An unanswered question is whether tissue-resident immune cells play a role in shaping the tissue as it develops in utero. Here, we profiled human embryonic and fetal lung immune cells using scRNA-seq, smFISH, and immunohistochemistry. At the embryonic stage, we observed an early wave of innate immune cells, including innate lymphoid cells, natural killer cells, myeloid cells, and lineage progenitors. By the canalicular stage, we detected naive T lymphocytes expressing high levels of cytotoxicity genes and the presence of mature B lymphocytes, including B-1 cells. Our analysis suggests that fetal lungs provide a niche for full B cell maturation. Given the presence and diversity of immune cells during development, we also investigated their possible effect on epithelial maturation. We found that IL-1ß drives epithelial progenitor exit from self-renewal and differentiation to basal cells in vitro. In vivo, IL-1ß-producing myeloid cells were found throughout the lung and adjacent to epithelial tips, suggesting that immune cells may direct human lung epithelial development.

Data Augmentation of X-ray Images for Automatic Cargo Inspection of Nuclear Items.

As part of establishing a management system to prevent the illegal transfer of nuclear items, automatic nuclear item detection technology is required during customs clearance. However, it is challenging to acquire X-ray images of major nuclear items (e.g., nuclear fuel and gas centrifuges) loaded in cargo with which to train a cargo inspection model. In this work, we propose a new means of data augmentation to alleviate the lack of X-ray training data. The proposed augmentation method generates synthetic X-ray images for the training of semantic segmentation models combining the X-ray images of nuclear items and X-ray cargo background images. To evaluate the effectiveness of the proposed data augmentation technique, we trained representative semantic segmentation models and performed extensive experiments to assess its quantitative and qualitative performance capabilities. Our findings show that multiple item insertions to respond to actual X-ray cargo inspection situations and the resulting occlusion expressions significantly affect the performance of the segmentation models. We believe that this augmentation research will enhance automatic cargo inspections to prevent the illegal transfer of nuclear items at airports and ports.

Lightweight image steganalysis with block-wise pruning.

Image steganalysis is the task of detecting a secret message hidden in an image. Deep steganalysis using end-to-end deep learning has been successful in recent years, but previous studies focused on improving detection performance rather than designing a lightweight model for practical applications. This caused a deep steganalysis model to be heavy and computationally costly, making the model infeasible to deploy in real-world applications. To address this issue, we study an effective model design strategy for lightweight image steganalysis. Considering the domain-specific characteristics of steganalysis, we propose a simple yet effective block removal strategy that progressively removes a sequence of blocks from deep classification networks. This method involves the gradual removal of convolutional neural network blocks, starting from deeper ones. By doing so, the number of parameters and FLOPs are decreased without compromising the detection performance. Experimental results show that our removal strategy makes the EfficientNet-B0 variants 9.58 [Formula: see text] smaller and has 2.16 [Formula: see text] fewer FLOPs than the baseline while retaining detection accuracy of 90.73% and 82.40% that are on par with the baseline on BOSSBase and ALASKA#2 datasets, respectively. Backed by our in-depth analyses, the results indicate that only a few early layers are sufficient for effective image steganalysis.

A novel human fetal lung-derived alveolar organoid model reveals mechanisms of surfactant protein C maturation relevant to interstitial lung disease.

Alveolar type 2 (AT2) cells maintain lung health by acting as stem cells and producing pulmonary surfactant1-3. AT2 dysfunction underlies many lung diseases including interstitial lung disease (ILD), in which some inherited forms result from mislocalisation of surfactant protein C (SFTPC) variants4,5. Disease modelling and dissection of mechanisms remains challenging due to complexities in deriving and maintaining AT2 cells ex vivo. Here, we describe the development of expandable adult AT2-like organoids derived from human fetal lung which are phenotypically stable, can differentiate into AT1-like cells and are genetically manipulable. We use these organoids to test key effectors of SFTPC maturation identified in a forward genetic screen including the E3 ligase ITCH, demonstrating that their depletion phenocopies the pathological SFTPC redistribution seen for the SFTPC-I73T variant. In summary, we demonstrate the development of a novel alveolar organoid model and use it to identify effectors of SFTPC maturation necessary for AT2 health.

Organoid modeling of human fetal lung alveolar development reveals mechanisms of cell fate patterning and neonatal respiratory disease.

Variation in lung alveolar development is strongly linked to disease susceptibility. However, underlying cellular and molecular mechanisms are difficult to study in humans. We have identified an alveolar-fated epithelial progenitor in human fetal lungs, which we grow as self-organizing organoids that model key aspects of cell lineage commitment. Using this system, we have functionally validated cell-cell interactions in the developing human alveolar niche, showing that Wnt signaling from differentiating fibroblasts promotes alveolar-type-2 cell identity, whereas myofibroblasts secrete the Wnt inhibitor, NOTUM, providing spatial patterning. We identify a Wnt-NKX2.1 axis controlling alveolar differentiation. Moreover, we show that differential binding of NKX2.1 coordinates alveolar maturation, allowing us to model the effects of human genetic variation in NKX2.1 on alveolar differentiation. Our organoid system recapitulates key aspects of human fetal lung stem cell biology allowing mechanistic experiments to determine the cellular and molecular regulation of human development and disease.

A human fetal lung cell atlas uncovers proximal-distal gradients of differentiation and key regulators of epithelial fates.

We present a multiomic cell atlas of human lung development that combines single-cell RNA and ATAC sequencing, high-throughput spatial transcriptomics, and single-cell imaging. Coupling single-cell methods with spatial analysis has allowed a comprehensive cellular survey of the epithelial, mesenchymal, endothelial, and erythrocyte/leukocyte compartments from 5-22 post-conception weeks. We identify previously uncharacterized cell states in all compartments. These include developmental-specific secretory progenitors and a subtype of neuroendocrine cell related to human small cell lung cancer. Our datasets are available through our web interface (https://lungcellatlas.org). To illustrate its general utility, we use our cell atlas to generate predictions about cell-cell signaling and transcription factor hierarchies which we rigorously test using organoid models.

SOX9 maintains human foetal lung tip progenitor state by enhancing WNT and RTK signalling.

The balance between self-renewal and differentiation in human foetal lung epithelial progenitors controls the size and function of the adult organ. Moreover, progenitor cell gene regulation networks are employed by both regenerating and malignant lung cells, where modulators of their effects could potentially be of therapeutic value. Details of the molecular networks controlling human lung progenitor self-renewal remain unknown. We performed the first CRISPRi screen in primary human lung organoids to identify transcription factors controlling progenitor self-renewal. We show that SOX9 promotes proliferation of lung progenitors and inhibits precocious airway differentiation. Moreover, by identifying direct transcriptional targets using Targeted DamID, we place SOX9 at the centre of a transcriptional network, which amplifies WNT and RTK signalling to stabilise the progenitor cell state. In addition, the proof-of-principle CRISPRi screen and Targeted DamID tools establish a new workflow for using primary human organoids to elucidate detailed functional mechanisms underlying normal development and disease.

A functional genetic toolbox for human tissue-derived organoids.

Human organoid systems recapitulate key features of organs offering platforms for modelling developmental biology and disease. Tissue-derived organoids have been widely used to study the impact of extrinsic niche factors on stem cells. However, they are rarely used to study endogenous gene function due to the lack of efficient gene manipulation tools. Previously, we established a human foetal lung organoid system (Nikolic et al., 2017). Here, using this organoid system as an example, we have systematically developed and optimised a complete genetic toolbox for use in tissue-derived organoids. This includes 'Organoid Easytag', our efficient workflow for targeting all types of gene loci through CRISPR-mediated homologous recombination followed by flow cytometry for enriching correctly targeted cells. Our toolbox also incorporates conditional gene knockdown or overexpression using tightly inducible CRISPR interference and CRISPR activation which is the first efficient application of these techniques to tissue-derived organoids. These tools will facilitate gene perturbation studies in tissue-derived organoids facilitating human disease modelling and providing a functional counterpart to many ongoing descriptive studies, such as the Human Cell Atlas Project.