Developing inhibitory peptides against SARS-CoV-2 envelope protein.

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has affected approximately 800 million people since the start of the Coronavirus Disease 2019 (COVID-19) pandemic. Because of the high rate of mutagenesis in SARS-CoV-2, it is difficult to develop a sustainable approach for prevention and treatment. The Envelope (E) protein is highly conserved among human coronaviruses. Previous studies reported that SARS-CoV-1 E deficiency reduced viral propagation, suggesting that E inhibition might be an effective therapeutic strategy for SARS-CoV-2. Here, we report inhibitory peptides against SARS-CoV-2 E protein named iPep-SARS2-E. Leveraging E-induced alterations in proton homeostasis and NFAT/AP-1 pathway in mammalian cells, we developed screening platforms to design and optimize the peptides that bind and inhibit E protein. Using Vero-E6 cells, human-induced pluripotent stem cell-derived branching lung organoid and mouse models with SARS-CoV-2, we found that iPep-SARS2-E significantly inhibits virus egress and reduces viral cytotoxicity and propagation in vitro and in vivo. Furthermore, the peptide can be customizable for E protein of other human coronaviruses such as Middle East Respiratory Syndrome Coronavirus (MERS-CoV). The results indicate that E protein can be a potential therapeutic target for human coronaviruses.

Health System Perspective on Cost for Delivering a Decision Aid for Prostate Cancer Using Time-driven Activity-based Costing.

BACKGROUND: Previsit decision aids (DAs) have promising outcomes in improving decisional quality, however, the cost to deploy a DA is not well defined, presenting a possible barrier to health system adoption. OBJECTIVES: We aimed to define the cost from a health system perspective of delivery of a DA. RESEARCH DESIGN: Observational cohort. PATIENTS AND METHODS: We interviewed or observed relevant personnel at 3 institutions with implemented DA distribution programs targeting men with prostate cancer. We then created process maps for DA delivery based on interview data. Cost determination was performed utilizing time-driven activity-based costing. Clinic visit length was measured on a subset of patients. Decisional quality measures were collected after the clinic visit. RESULTS: Total process time (minutes) for DA delivery was 10.14 (UCLA), 68 (Olive View-UCLA), and 25 (Vanderbilt). Total average costs (USD) per patient were $38.32 (UCLA), $59.96 (Olive View-UCLA), and $42.38 (Vanderbilt), respectively. Labor costs were the largest contributors to the cost of DA delivery. Variance analyses confirmed the cost efficiency of electronic health record (EHR) integration. We noted a shortening of clinic visit length when the DA was used, with high levels of decision quality. CONCLUSIONS: Time-driven activity-based costing is an effective approach to determining true inclusive costs of service delivery while also elucidating opportunities for cost containment. The absolute cost of delivering a DA to men with prostate cancer in various settings is much lower than the system costs of the treatments they consider. EHR integration streamlines DA delivery efficiency and results in substantial cost savings.

PAIP 2020: Microsatellite instability prediction in colorectal cancer.

Microsatellite instability (MSI) refers to alterations in the length of simple repetitive genomic sequences. MSI status serves as a prognostic and predictive factor in colorectal cancer. The MSI-high status is a good prognostic factor in stage II/III cancer, and predicts a lack of benefit to adjuvant fluorouracil chemotherapy in stage II cancer but a good response to immunotherapy in stage IV cancer. Therefore, determining MSI status in patients with colorectal cancer is important for identifying the appropriate treatment protocol. In the Pathology Artificial Intelligence Platform (PAIP) 2020 challenge, artificial intelligence researchers were invited to predict MSI status based on colorectal cancer slide images. Participants were required to perform two tasks. The primary task was to classify a given slide image as belonging to either the MSI-high or the microsatellite-stable group. The second task was tumor area segmentation to avoid ties with the main task. A total of 210 of the 495 participants enrolled in the challenge downloaded the images, and 23 teams submitted their final results. Seven teams from the top 10 participants agreed to disclose their algorithms, most of which were convolutional neural network-based deep learning models, such as EfficientNet and UNet. The top-ranked system achieved the highest F1 score (0.9231). This paper summarizes the various methods used in the PAIP 2020 challenge. This paper supports the effectiveness of digital pathology for identifying the relationship between colorectal cancer and the MSI characteristics.

Deep Learning-Based Objective and Reproducible Osteosarcoma Chemotherapy Response Assessment and Outcome Prediction.

Osteosarcoma is the most common primary bone cancer, whose standard treatment includes pre-operative chemotherapy followed by resection. Chemotherapy response is used for prognosis and management of patients. Necrosis is routinely assessed after chemotherapy from histology slides on resection specimens, where necrosis ratio is defined as the ratio of necrotic tumor/overall tumor. Patients with necrosis ratio ≥90% are known to have a better outcome. Manual microscopic review of necrosis ratio from multiple glass slides is semiquantitative and can have intraobserver and interobserver variability. In this study, an objective and reproducible deep learning-based approach was proposed to estimate necrosis ratio with outcome prediction from scanned hematoxylin and eosin whole slide images (WSIs). To conduct the study, 103 osteosarcoma cases with 3134 WSIs were collected. Deep Multi-Magnification Network was trained to segment multiple tissue subtypes, including viable tumor and necrotic tumor at a pixel level and to calculate case-level necrosis ratio from multiple WSIs. Necrosis ratio estimated by the segmentation model highly correlates with necrosis ratio from pathology reports manually assessed by experts. Furthermore, patients were successfully stratified to predict overall survival with P = 2.4 × 10-6 and progression-free survival with P = 0.016. This study indicates that deep learning can support pathologists as an objective tool to analyze osteosarcoma from histology for assessing treatment response and predicting patient outcome.

Deep Interactive Learning-based ovarian cancer segmentation of H&E-stained whole slide images to study morphological patterns of BRCA mutation.

Deep learning has been widely used to analyze digitized hematoxylin and eosin (H&E)-stained histopathology whole slide images. Automated cancer segmentation using deep learning can be used to diagnose malignancy and to find novel morphological patterns to predict molecular subtypes. To train pixel-wise cancer segmentation models, manual annotation from pathologists is generally a bottleneck due to its time-consuming nature. In this paper, we propose Deep Interactive Learning with a pretrained segmentation model from a different cancer type to reduce manual annotation time. Instead of annotating all pixels from cancer and non-cancer regions on giga-pixel whole slide images, an iterative process of annotating mislabeled regions from a segmentation model and training/finetuning the model with the additional annotation can reduce the time. Especially, employing a pretrained segmentation model can further reduce the time than starting annotation from scratch. We trained an accurate ovarian cancer segmentation model with a pretrained breast segmentation model by 3.5 hours of manual annotation which achieved intersection-over-union of 0.74, recall of 0.86, and precision of 0.84. With automatically extracted high-grade serous ovarian cancer patches, we attempted to train an additional classification deep learning model to predict BRCA mutation. The segmentation model and code have been released at https://github.com/MSKCC-Computational-Pathology/DMMN-ovary.

Mapping monoclonal anti-SARS-CoV-2 antibody repertoires against diverse coronavirus antigens.

Variants of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have emerged continuously, challenging the effectiveness of vaccines, diagnostics, and treatments. Moreover, the possibility of the appearance of a new betacoronavirus with high transmissibility and high fatality is reason for concern. In this study, we used a natively paired yeast display technology, combined with next-generation sequencing (NGS) and massive bioinformatic analysis to perform a comprehensive study of subdomain specificity of natural human antibodies from two convalescent donors. Using this screening technology, we mapped the cross-reactive responses of antibodies generated by the two donors against SARS-CoV-2 variants and other betacoronaviruses. We tested the neutralization potency of a set of the cross-reactive antibodies generated in this study and observed that most of the antibodies produced by these patients were non-neutralizing. We performed a comparison of the specific and non-specific antibodies by somatic hypermutation in a repertoire-scale for the two individuals and observed that the degree of somatic hypermutation was unique for each patient. The data from this study provide functional insights into cross-reactive antibodies that can assist in the development of strategies against emerging SARS-CoV-2 variants and divergent betacoronaviruses.

Functional map of SARS-CoV-2 3CL protease reveals tolerant and immutable sites.

The SARS-CoV-2 3CL protease (3CLpro) is an attractive therapeutic target, as it is essential to the virus and highly conserved among coronaviruses. However, our current understanding of its tolerance to mutations is limited. Here, we develop a yeast-based deep mutational scanning approach to systematically profile the activity of all possible single mutants of the 3CLpro and validate a subset of our results within authentic viruses. We reveal that the 3CLpro is highly malleable and is capable of tolerating mutations throughout the protein. Yet, we also identify specific residues that appear immutable, suggesting that these may be targets for future 3CLpro inhibitors. Finally, we utilize our screening as a basis to identify E166V as a resistance-conferring mutation against the clinically used 3CLpro inhibitor, nirmatrelvir. Collectively, the functional map presented herein may serve as a guide to better understand the biological properties of the 3CLpro and for drug development against coronaviruses.

A putative cap binding protein and the methyl phosphate capping enzyme Bin3/MePCE function in telomerase biogenesis.

Telomerase reverse transcriptase (TERT) and the noncoding telomerase RNA (TR) subunit constitute the core of telomerase. Additional subunits are required for ribonucleoprotein complex assembly and in some cases remain stably associated with the active holoenzyme. Pof8, a member of the LARP7 protein family is such a constitutive component of telomerase in fission yeast. Using affinity purification of Pof8, we have identified two previously uncharacterized proteins that form a complex with Pof8 and participate in telomerase biogenesis. Both proteins participate in ribonucleoprotein complex assembly and are required for wildtype telomerase activity and telomere length maintenance. One factor we named Thc1 (Telomerase Holoenzyme Component 1) shares structural similarity with the nuclear cap binding complex and the poly-adenosine ribonuclease (PARN), the other is the ortholog of the methyl phosphate capping enzyme (Bin3/MePCE) in metazoans and was named Bmc1 (Bin3/MePCE 1) to reflect its evolutionary roots. Thc1 and Bmc1 function together with Pof8 in recognizing correctly folded telomerase RNA and promoting the recruitment of the Lsm2-8 complex and the catalytic subunit to assemble functional telomerase.

An airway organoid-based screen identifies a role for the HIF1α-glycolysis axis in SARS-CoV-2 infection.

It is urgent to develop disease models to dissect mechanisms regulating severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. Here, we derive airway organoids from human pluripotent stem cells (hPSC-AOs). The hPSC-AOs, particularly ciliated-like cells, are permissive to SARS-CoV-2 infection. Using this platform, we perform a high content screen and identify GW6471, which blocks SARS-CoV-2 infection. GW6471 can also block infection of the B.1.351 SARS-CoV-2 variant. RNA sequencing (RNA-seq) analysis suggests that GW6471 blocks SARS-CoV-2 infection at least in part by inhibiting hypoxia inducible factor 1 subunit alpha (HIF1α), which is further validated by chemical inhibitor and genetic perturbation targeting HIF1α. Metabolic profiling identifies decreased rates of glycolysis upon GW6471 treatment, consistent with transcriptome profiling. Finally, xanthohumol, 5-(tetradecyloxy)-2-furoic acid, and ND-646, three compounds that suppress fatty acid biosynthesis, also block SARS-CoV-2 infection. Together, a high content screen coupled with transcriptome and metabolic profiling reveals a key role of the HIF1α-glycolysis axis in mediating SARS-CoV-2 infection of human airway epithelium.

Cardiomyocytes recruit monocytes upon SARS-CoV-2 infection by secreting CCL2.

Heart injury has been reported in up to 20% of COVID-19 patients, yet the cause of myocardial histopathology remains unknown. Here, using an established in vivo hamster model, we demonstrate that SARS-CoV-2 can be detected in cardiomyocytes of infected animals. Furthermore, we found damaged cardiomyocytes in hamsters and COVID-19 autopsy samples. To explore the mechanism, we show that both human pluripotent stem cell-derived cardiomyocytes (hPSC-derived CMs) and adult cardiomyocytes (CMs) can be productively infected by SARS-CoV-2, leading to secretion of the monocyte chemoattractant cytokine CCL2 and subsequent monocyte recruitment. Increased CCL2 expression and monocyte infiltration was also observed in the hearts of infected hamsters. Although infected CMs suffer damage, we find that the presence of macrophages significantly reduces SARS-CoV-2-infected CMs. Overall, our study provides direct evidence that SARS-CoV-2 infects CMs in vivo and suggests a mechanism of immune cell infiltration and histopathology in heart tissues of COVID-19 patients.

Multi-magnification-based machine learning as an ancillary tool for the pathologic assessment of shaved margins for breast carcinoma lumpectomy specimens.

The surgical margin status of breast lumpectomy specimens for invasive carcinoma and ductal carcinoma in situ (DCIS) guides clinical decisions, as positive margins are associated with higher rates of local recurrence. The "cavity shave" method of margin assessment has the benefits of allowing the surgeon to orient shaved margins intraoperatively and the pathologist to assess one inked margin per specimen. We studied whether a deep convolutional neural network, a deep multi-magnification network (DMMN), could accurately segment carcinoma from benign tissue in whole slide images (WSIs) of shave margin slides, and therefore serve as a potential screening tool to improve the efficiency of microscopic evaluation of these specimens. Applying the pretrained DMMN model, or the initial model, to a validation set of 408 WSIs (348 benign, 60 with carcinoma) achieved an area under the curve (AUC) of 0.941. After additional manual annotations and fine-tuning of the model, the updated model achieved an AUC of 0.968 with sensitivity set at 100% and corresponding specificity of 78%. We applied the initial model and updated model to a testing set of 427 WSIs (374 benign, 53 with carcinoma) which showed AUC values of 0.900 and 0.927, respectively. Using the pixel classification threshold selected from the validation set, the model achieved a sensitivity of 92% and specificity of 78%. The four false-negative classifications resulted from two small foci of DCIS (1 mm, 0.5 mm) and two foci of well-differentiated invasive carcinoma (3 mm, 1.5 mm). This proof-of-principle study demonstrates that a DMMN machine learning model can segment invasive carcinoma and DCIS in surgical margin specimens with high accuracy and has the potential to be used as a screening tool for pathologic assessment of these specimens.

An Immuno-Cardiac Model for Macrophage-Mediated Inflammation in COVID-19 Hearts.

[Figure: see text].

Inhibitors of Coronavirus 3CL Proteases Protect Cells from Protease-Mediated Cytotoxicity.

We describe a mammalian cell-based assay to identify coronavirus 3CL protease (3CLpro) inhibitors. This assay is based on rescuing protease-mediated cytotoxicity and does not require live virus. By enabling the facile testing of compounds across a range of 15 distantly related coronavirus 3CLpro enzymes, we identified compounds with broad 3CLpro-inhibitory activity. We also adapted the assay for use in compound screening and in doing so uncovered additional severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 3CLpro inhibitors. We observed strong concordance between data emerging from this assay and those obtained from live-virus testing. The reported approach democratizes the testing of 3CLpro inhibitors by developing a simplified method for identifying coronavirus 3CLpro inhibitors that can be used by the majority of laboratories, rather than the few with extensive biosafety infrastructure. We identified two lead compounds, GC376 and compound 4, with broad activity against all 3CL proteases tested, including 3CLpro enzymes from understudied zoonotic coronaviruses. IMPORTANCE Multiple coronavirus pandemics have occurred over the last 2 decades. This has highlighted a need to be proactive in the development of therapeutics that can be readily deployed in the case of future coronavirus pandemics. We developed and validated a simplified cell-based assay for the identification of chemical inhibitors of 3CL proteases encoded by a wide range of coronaviruses. This assay is reporter free, does not require specialized biocontainment, and is optimized for performance in high-throughput screening. By testing reported 3CL protease inhibitors against a large collection of 3CL proteases with variable sequence similarity, we identified compounds with broad activity against 3CL proteases and uncovered structural insights into features that contribute to their broad activity. Furthermore, we demonstrated that this assay is suitable for identifying chemical inhibitors of proteases from families other than 3CL proteases.

Deep Multi-Magnification Networks for multi-class breast cancer image segmentation.

Pathologic analysis of surgical excision specimens for breast carcinoma is important to evaluate the completeness of surgical excision and has implications for future treatment. This analysis is performed manually by pathologists reviewing histologic slides prepared from formalin-fixed tissue. In this paper, we present Deep Multi-Magnification Network trained by partial annotation for automated multi-class tissue segmentation by a set of patches from multiple magnifications in digitized whole slide images. Our proposed architecture with multi-encoder, multi-decoder, and multi-concatenation outperforms other single and multi-magnification-based architectures by achieving the highest mean intersection-over-union, and can be used to facilitate pathologists' assessments of breast cancer.

SARS-CoV-2 neutralizing antibody responses are more robust in patients with severe disease.

We studied plasma antibody responses of 35 patients about 1 month after SARS-CoV-2 infection. Titers of antibodies binding to the viral nucleocapsid and spike proteins were significantly higher in patients with severe disease. Likewise, mean antibody neutralization titers against SARS-CoV-2 pseudovirus and live virus were higher in the sicker patients, by ∼5-fold and ∼7-fold, respectively. These findings have important implications for those pursuing plasma therapy, isolation of neutralizing monoclonal antibodies, and determinants of immunity.

CRISPR-based gene knockout screens reveal deubiquitinases involved in HIV-1 latency in two Jurkat cell models.

The major barrier to a HIV-1 cure is the persistence of latent genomes despite treatment with antiretrovirals. To investigate host factors which promote HIV-1 latency, we conducted a genome-wide functional knockout screen using CRISPR-Cas9 in a HIV-1 latency cell line model. This screen identified IWS1, POLE3, POLR1B, PSMD1, and TGM2 as potential regulators of HIV-1 latency, of which PSMD1 and TMG2 could be confirmed pharmacologically. Further investigation of PSMD1 revealed that an interacting enzyme, the deubiquitinase UCH37, was also involved in HIV-1 latency. We therefore conducted a comprehensive evaluation of the deubiquitinase family by gene knockout, identifying several deubiquitinases, UCH37, USP14, OTULIN, and USP5 as possible HIV-1 latency regulators. A specific inhibitor of USP14, IU1, reversed HIV-1 latency and displayed synergistic effects with other latency reversal agents. IU1 caused degradation of TDP-43, a negative regulator of HIV-1 transcription. Collectively, this study is the first comprehensive evaluation of deubiquitinases in HIV-1 latency and establishes that they may hold a critical role.

The Miami Monkey: A Sunny Alternative to the Berlin Patient.

Curing HIV infection has been impossible, with the exception of the "Berlin Patient." Martinez-Navio et al. (2019) in Miami herein present a rare monkey whose virus was controlled for >3 years after a single genetic intervention that led to persistent production of HIV-neutralizing antibodies in vivo.

Increasing Awareness of the Importance of Physical Activity and Healthy Nutrition: Results From a Mixed-Methods Evaluation of a Workplace Program.

BACKGROUND: To evaluate the impact of an online workplace program that promotes physical activity and health, while focusing on performance measures relating to physical activity, nutrition, and overall health. METHODS: The large sample size (more than 18,000 participants) allowed the use of text mining and machine-learning methods to determine what descriptions of the program identify successful outcomes, and hierarchical linear models to determine the most beneficial program modules and features. RESULTS: The program increased overall health and awareness of levels of physical activity and nutrition, especially for people who scored low on these measures initially. Interestingly, although physical activity is the most popular program module, the daily step-tracking process was associated with smaller improvements in overall health. CONCLUSIONS: This study finds that the Virgin Pulse Global Challenge is an effective workplace intervention for improving overall health and awareness of physical activity and nutrition. Effectiveness relates to the holistic approach adopted rather than to individual modules in isolation. Future evaluations of workplace health and exercise programs should explore a variety of outcome measures within the rich context provided by open-ended participant experience feedback. In addition, a control group and a follow-up study are required.