MCA-UNet: multi-scale cross co-attentional U-Net for automatic medical image segmentation.

Medical image segmentation is a challenging task due to the high variation in shape, size and position of infections or lesions in medical images. It is necessary to construct multi-scale representations to capture image contents from different scales. However, it is still challenging for U-Net with a simple skip connection to model the global multi-scale context. To overcome it, we proposed a dense skip-connection with cross co-attention in U-Net to solve the semantic gaps for an accurate automatic medical image segmentation. We name our method MCA-UNet, which enjoys two benefits: (1) it has a strong ability to model the multi-scale features, and (2) it jointly explores the spatial and channel attentions. The experimental results on the COVID-19 and IDRiD datasets suggest that our MCA-UNet produces more precise segmentation performance for the consolidation, ground-glass opacity (GGO), microaneurysms (MA) and hard exudates (EX). The source code of this work will be released via https://github.com/McGregorWwww/MCA-UNet/.

Human carboxylesterase 1A plays a predominant role in the hydrolytic activation of remdesivir in humans.

Remdesivir, an intravenous nucleotide prodrug, has been approved for treating COVID-19 in hospitalized adults and pediatric patients. Upon administration, remdesivir can be readily hydrolyzed to form its active form GS-441524, while the cleavage of the carboxylic ester into GS-704277 is the first step for remdesivir activation. This study aims to assign the key enzymes responsible for remdesivir hydrolysis in humans, as well as to investigate the kinetics of remdesivir hydrolysis in various enzyme sources. The results showed that remdesivir could be hydrolyzed to form GS-704277 in human plasma and the microsomes from human liver (HLMs), lung (HLuMs) and kidney (HKMs), while the hydrolytic rate of remdesivir in HLMs was the fastest. Chemical inhibition and reaction phenotyping assays suggested that human carboxylesterase 1 (hCES1A) played a predominant role in remdesivir hydrolysis, while cathepsin A (CTSA), acetylcholinesterase (AchE) and butyrylcholinesterase (BchE) contributed to a lesser extent. Enzymatic kinetic analyses demonstrated that remdesivir hydrolysis in hCES1A (SHUTCM) and HLMs showed similar kinetic plots and much closed Km values to each other. Meanwhile, GS-704277 formation rates were strongly correlated with the CES1A activities in HLM samples from different individual donors. Further investigation revealed that simvastatin (a therapeutic agent for adjuvant treating COVID-19) strongly inhibited remdesivir hydrolysis in both recombinant hCES1A and HLMs. Collectively, our findings reveal that hCES1A plays a predominant role in remdesivir hydrolysis in humans, which are very helpful for predicting inter-individual variability in response to remdesivir and for guiding the rational use of this anti-COVID-19 agent in clinical settings.

Flavonoids in Ampelopsis grossedentata as covalent inhibitors of SARS-CoV-2 3CLpro: Inhibition potentials, covalent binding sites and inhibitory mechanisms.

Coronavirus 3C-like protease (3CLpro) is a crucial target for treating coronavirus diseases including COVID-19. Our preliminary screening showed that Ampelopsis grossedentata extract (AGE) displayed potent SARS-CoV-2-3CLpro inhibitory activity, but the key constituents with SARS-CoV-2-3CLpro inhibitory effect and their mechanisms were unrevealed. Herein, a practical strategy via integrating bioactivity-guided fractionation and purification, mass spectrometry-based peptide profiling and time-dependent biochemical assay, was applied to identify the crucial constituents in AGE and to uncover their inhibitory mechanisms. The results demonstrated that the flavonoid-rich fractions (10-17.5 min) displayed strong SARS-CoV-2-3CLpro inhibitory activities, while the constituents in these fractions were isolated and their SARS-CoV-2-3CLpro inhibitory activities were investigated. Among all isolated flavonoids, dihydromyricetin, isodihydromyricetin and myricetin strongly inhibited SARS-CoV-2 3CLpro in a time-dependent manner. Further investigations demonstrated that myricetin could covalently bind on SARS-CoV-2 3CLpro at Cys300 and Cys44, while dihydromyricetin and isodihydromyricetin covalently bound at Cys300. Covalent docking coupling with molecular dynamics simulations showed the detailed interactions between the orthoquinone form of myricetin and two covalent binding sites (surrounding Cys300 and Cys44) of SARS-CoV-2 3CLpro. Collectively, the flavonoids in AGE strongly and time-dependently inhibit SARS-CoV-2 3CLpro, while the newly identified SARS-CoV-2 3CLpro inhibitors in AGE offer promising lead compounds for developing novel antiviral agents.

Identification of Vitamin K3 and its analogues as covalent inhibitors of SARS-CoV-2 3CLpro.

After the emergence of the pandemic, repurposed drugs have been considered as a quicker way of finding potential antiviral agents. SARS-CoV-2 3CLpro is essential for processing the viral polyproteins into mature non-structural proteins, making it an attractive target for developing antiviral agents. Here we show that Vitamin K3 screened from the FDA-Approved Drug Library containing an array of 1,018 compounds has potent inhibitory activity against SARS-CoV-2 3CLpro with the IC50 value of 4.78 ± 1.03 µM, rather than Vitamin K1, K2 and K4. Next, the time-dependent inhibitory experiment was carried out to confirm that Vitamin K3 could form the covalent bond with SARS-CoV-2 3CLpro. Then we analyzed the structure-activity relationship of Vitamin K3 analogues and identified 5,8-dihydroxy-1,4-naphthoquinone with 9.8 times higher inhibitory activity than Vitamin K3. Further mass spectrometric analysis and molecular docking study verified the covalent binding between Vitamin K3 or 5,8-dihydroxy-1,4-naphthoquinone and SARS-CoV-2 3CLpro. Thus, our findings provide valuable information for further optimization and design of novel inhibitors based on Vitamin K3 and its analogues, which may have the potential to fight against SARS-CoV-2.

Discovery of naturally occurring inhibitors against SARS-CoV-2 3CLpro from Ginkgo biloba leaves via large-scale screening.

3-Chymotrypsin-like protease (3CLpro) is a virally encoded main proteinase that is pivotal for the viral replication across a broad spectrum of coronaviruses. This study aims to discover the naturally occurring SARS-CoV-2 3CLpro inhibitors from herbal constituents, as well as to investigate the inhibitory mechanism of the newly identified efficacious SARS-CoV-2 3CLpro inhibitors. Following screening of the inhibitory potentials of eighty herbal products against SARS-CoV-2 3CLpro, Ginkgo biloba leaves extract (GBLE) was found with the most potent SARS-CoV-2 3CLpro inhibition activity (IC50 = 6.68 µg/mL). Inhibition assays demonstrated that the ginkgolic acids (GAs) and the bioflavones isolated from GBLE displayed relatively strong SARS-CoV-2 3CLpro inhibition activities (IC50 < 10 µM). Among all tested constituents, GA C15:0, GA C17:1 and sciadopitysin displayed potent 3CLpro inhibition activities, with IC50 values of less than 2 µM. Further inhibition kinetic studies and docking simulations clearly demonstrated that two GAs and sciadopitysin strongly inhibit SARS-CoV-2 3CLprovia a reversible and mixed inhibition manner. Collectively, this study found that both GBLE and the major constituents in this herbal product exhibit strong SARS-CoV-2 3CLpro inhibition activities, which offer several promising leading compounds for developing novel anti-COVID-19 medications via targeting on 3CLpro.