An explainable machine learning-driven proposal of pulmonary fibrosis biomarkers.

Pulmonary fibrosing diseases are in the very epicenter of biomedical research both due to their increasing prevalence and their association with SARS-CoV-2 infections. Research of idiopathic pulmonary fibrosis, the most lethal among the interstitial lung diseases, is in need for new biomarkers and potential disease targets, a goal that could be accelerated using machine learning techniques. In this study, we have used Shapley values to explain the decisions made by an ensemble learning model trained to classify samples to an either pulmonary fibrosis or steady state based on the expression values of deregulated genes. This process resulted in a full and a laconic set of features capable of separating phenotypes to an at least equal degree as previously published marker sets. Indicatively, a maximum increase of 6% in specificity and 5% in Mathew's correlation coefficient was achieved. Evaluation with an additional independent dataset showed our feature set having a greater generalization potential than the rest. Ultimately, the proposed gene lists are expected not only to serve as new sets of diagnostic marker elements, but also as a target pool for future research initiatives.

Commonalities Between ARDS, Pulmonary Fibrosis and COVID-19: The Potential of Autotaxin as a Therapeutic Target.

Severe COVID-19 is characterized by acute respiratory distress syndrome (ARDS)-like hyperinflammation and endothelial dysfunction, that can lead to respiratory and multi organ failure and death. Interstitial lung diseases (ILD) and pulmonary fibrosis confer an increased risk for severe disease, while a subset of COVID-19-related ARDS surviving patients will develop a fibroproliferative response that can persist post hospitalization. Autotaxin (ATX) is a secreted lysophospholipase D, largely responsible for the extracellular production of lysophosphatidic acid (LPA), a pleiotropic signaling lysophospholipid with multiple effects in pulmonary and immune cells. In this review, we discuss the similarities of COVID-19, ARDS and ILDs, and suggest ATX as a possible pathologic link and a potential common therapeutic target.

Increased Autotaxin Levels in Severe COVID-19, Correlating with IL-6 Levels, Endothelial Dysfunction Biomarkers, and Impaired Functions of Dendritic Cells.

Autotaxin (ATX; ENPP2) is a secreted lysophospholipase D catalyzing the extracellular production of lysophosphatidic acid (LPA), a pleiotropic signaling phospholipid. Genetic and pharmacologic studies have previously established a pathologic role for ATX and LPA signaling in pulmonary injury, inflammation, and fibrosis. Here, increased ENPP2 mRNA levels were detected in immune cells from nasopharyngeal swab samples of COVID-19 patients, and increased ATX serum levels were found in severe COVID-19 patients. ATX serum levels correlated with the corresponding increased serum levels of IL-6 and endothelial damage biomarkers, suggesting an interplay of the ATX/LPA axis with hyperinflammation and the associated vascular dysfunction in COVID-19. Accordingly, dexamethasone (Dex) treatment of mechanically ventilated patients reduced ATX levels, as shown in two independent cohorts, indicating that the therapeutic benefits of Dex include the suppression of ATX. Moreover, large scale analysis of multiple single cell RNA sequencing datasets revealed the expression landscape of ENPP2 in COVID-19 and further suggested a role for ATX in the homeostasis of dendritic cells, which exhibit both numerical and functional deficits in COVID-19. Therefore, ATX has likely a multifunctional role in COVID-19 pathogenesis, suggesting that its pharmacological targeting might represent an additional therapeutic option, both during and after hospitalization.

Alteration of L-Dopa decarboxylase expression in SARS-CoV-2 infection and its association with the interferon-inducible ACE2 isoform.

L-Dopa decarboxylase (DDC) is the most significantly co-expressed gene with ACE2, which encodes for the SARS-CoV-2 receptor angiotensin-converting enzyme 2 and the interferon-inducible truncated isoform dACE2. Our group previously showed the importance of DDC in viral infections. We hereby aimed to investigate DDC expression in COVID-19 patients and cultured SARS-CoV-2-infected cells, also in association with ACE2 and dACE2. We concurrently evaluated the expression of the viral infection- and interferon-stimulated gene ISG56 and the immune-modulatory, hypoxia-regulated gene EPO. Viral load and mRNA levels of DDC, ACE2, dACE2, ISG56 and EPO were quantified by RT-qPCR in nasopharyngeal swab samples from COVID-19 patients, showing no or mild symptoms, and from non-infected individuals. Samples from influenza-infected patients were analyzed in comparison. SARS-CoV-2-mediated effects in host gene expression were validated in cultured virus-permissive epithelial cells. We found substantially higher gene expression of DDC in COVID-19 patients (7.6-fold; p = 1.2e-13) but not in influenza-infected ones, compared to non-infected subjects. dACE2 was more elevated (2.9-fold; p = 1.02e-16) than ACE2 (1.7-fold; p = 0.0005) in SARS-CoV-2-infected individuals. ISG56 (2.5-fold; p = 3.01e-6) and EPO (2.6-fold; p = 2.1e-13) were also increased. Detected differences were not attributed to enrichment of specific cell populations in nasopharyngeal tissue. While SARS-CoV-2 virus load was positively associated with ACE2 expression (r≥0.8, p<0.001), it negatively correlated with DDC, dACE2 (r≤-0.7, p<0.001) and EPO (r≤-0.5, p<0.05). Moreover, a statistically significant correlation between DDC and dACE2 expression was observed in nasopharyngeal swab and whole blood samples of both COVID-19 and non-infected individuals (r≥0.7). In VeroE6 cells, SARS-CoV-2 negatively affected DDC, ACE2, dACE2 and EPO mRNA levels, and induced cell death, while ISG56 was enhanced at early hours post-infection. Thus, the regulation of DDC, dACE2 and EPO expression in the SARS-CoV-2-infected nasopharyngeal tissue is possibly related with an orchestrated antiviral response of the infected host as the virus suppresses these genes to favor its propagation.