Mask removal : Face inpainting via attributes.

Due to the outbreak of the COVID-19 pandemic, wearing masks in public areas has become an effective way to slow the spread of disease. However, it also brings some challenges to applications in daily life as half of the face is occluded. Therefore, the idea of removing masks by face inpainting appeared. Face inpainting has achieved promising performance but always fails to guarantee high-fidelity. In this paper, we present a novel mask removal inpainting network based on face attributes known in advance including nose, chubby, makeup, gender, mouth, beard and young, aiming to ensure the repaired face image is closer to ground truth. To achieve this, a dual pipeline network based on GANs has been proposed, one of which is a reconstructive path used in training that utilizes missing regions in ground truth to get prior distribution, while the other is a generative path for predicting information in the masked region. To establish the process of mask removal, we build a synthetic facial occlusion that mimics the real mask. Experiments show that our method not only generates faces more similarly aligned with real attributes, but also ensures semantic and structural rationality compared with state-of-the-art methods.

Coupled CRC 2D and ALI 3D Cultures Express Receptors of Emerging Viruses and Are More Suitable for the Study of Viral Infections Compared to Conventional Cell Lines.

Infections of emerging and reemerging viruses (SARS-CoVs, influenza H1N1, etc.) largely and globally affect human health. Animal models often fail to reflect a physiological status because of species tropism of virus infection. Conventional cell lines are usually genetically and phenotypically different from primary cells. Developing an in vitro physiological model to study the infection of emerging viruses will facilitate our understanding of virus-host cell interactions, thereby benefiting antiviral drug discovery. In the current work, we first established normal airway epithelial cells (upper and lower airway track) in 2D and 3D culture systems using conditional reprogramming (CR) and air-liquid interface (ALI) techniques. These long-term cultures maintained differentiation potential. More importantly, these cells express two types of influenza virus receptors, α2-6-Gal- and α2-3-Gal-linked sialic acids, and angiotensin-converting enzyme 2 (ACE2), a receptor for SARS-CoVs as well. These cells were permissive to the infection of pandemic influenza H1N1 (H1N1pdm). In contrast, the lung cancer cell line A549 and immortalized airway epithelial cells (16HBE) were not susceptible to H1N1 infection. A virus-induced cytopathic effect (CPE) on 2D CRC cultures developed in a time-dependent manner. The pathological effects were also readily observed spreading from the apical layer to the basal layer of the 3D ALI culture. This integrated 2D CRC and 3D ALI cultures provide a physiological and personalized in vitro model to study the infection of emerging viruses. This novel model can be used for studying virus biology and host response to viral infection and for antiviral drug discovery.

Long Term Culture of Human Kidney Proximal Tubule Epithelial Cells Maintains Lineage Functions and Serves as an Ex vivo Model for Coronavirus Associated Kidney Injury.

The mechanism of how SARS-CoV-2 causes severe multi-organ failure is largely unknown. Acute kidney injury (AKI) is one of the frequent organ damage in severe COVID-19 patients. Previous studies have shown that human renal tubule cells could be the potential host cells targeted by SARS-CoV-2. Traditional cancer cell lines or immortalized cell lines are genetically and phenotypically different from host cells. Animal models are widely used, but often fail to reflect a physiological and pathogenic status because of species tropisms. There is an unmet need for normal human epithelial cells for disease modeling. In this study, we successfully established long term cultures of normal human kidney proximal tubule epithelial cells (KPTECs) in 2D and 3D culture systems using conditional reprogramming (CR) and organoids techniques. These cells had the ability to differentiate and repair DNA damage, and showed no transforming property. Importantly, the CR KPTECs maintained lineage function with expression of specific transporters (SLC34A3 and cubilin). They also expressed angiotensin-converting enzyme 2 (ACE2), a receptor for SARS-CoV and SARS-CoV-2. In contrast, cancer cell line did not express endogenous SLC34A3, cubilin and ACE2. Very interestingly, ACE2 expression was around twofold higher in 3D organoids culture compared to that in 2D CR culture condition. Pseudovirion assays demonstrated that SARS-CoV spike (S) protein was able to enter CR cells with luciferase reporter. This integrated 2D CR and 3D organoid cultures provide a physiological ex vivo model to study kidney functions, innate immune response of kidney cells to viruses, and a novel platform for drug discovery and safety evaluation.