Deficiency in platelet 12-lipoxygenase exacerbates inflammation and disease severity during SARS-CoV-2 infection

Abstract

Platelets, known for maintaining blood balance, also participate in antimicrobial defense. Upon severeacute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, platelets become hyperactivated, releasing molecules such as cytokines, granule contents, and bioactive lipids. The key effector biolipids produced by platelets include 12-hydroxyeicosatetraenoic acid (12-HETE) and 12-hydroxyeicosatrienoic acid (12-HETrE), produced by 12-lipoxygenase (12-LOX), and prostaglandins and thromboxane, produced by cyclooxygenase-1. While prostaglandin E2 and thromboxane B2 were previously associated with lung inflammation in severe COVID-19, the role of platelet 12-LOX in SARS-CoV-2 infection remains unclear. Using mice deficient for platelets' 12-LOX, we report that SARS-CoV-2 infection resulted in higher lung inflammation characterized by histopathological tissue analysis, increased leukocyte infiltrates, and cytokine production relative to wild-type mice. In addition, distinct platelet and lung transcriptomic changes, including alterations in NOD-like receptor (NLR) family pyrin domain-containing 1 (NLRP1) inflammasome-related gene expression, were observed. Mass spectrometry lipidomic analysis in 12-LOX-deficient-infected mice revealed significant changes in bioactive lipid content, including reduced levels of 12-HETrE that inversely correlated with disease severity. Finally, platelet 12-LOX deficiency was associated with increased morbidity and lower survival rates relative to wild type (WT) mice. Overall, this study highlights the complex interplay between 12-LOX-related lipid metabolism and inflammatory responses during SARS-CoV-2 infection. The findings provide valuable insights into potential therapeutic targets aimed at mitigating severe outcomes, emphasizing the pivotal role of platelet enzymes in the host response to viral infections.

Keywords: 12-HETrE; Inflammation; SARS-CoV-2; platelet type-12 lipoxygenase; platelets.

Fig. 1.

Activation of the cPLA₂α downstream pathway during SARS-CoV-2 infection. (A) Study design. Human washed platelet responses to A549-hACE2 cell culture conditioned media from mock-infected (C.Med) or infected SARS-CoV-2 (CoV-2) (cell culture supernatant obtained after 4 d of culture and treatment with b-propiolactone to inactivate the virus). (B) Platelets from four donors were stimulated for 60 min, labeled, fixed, and analyzed by flow cytometry. (C) Lipid mediators, 12-HETE;12-HETrE; 12-HEPE; PGE_2; TXB₂ produced by human platelet in response to SARS-CoV-2 exposure were analyzed by LC–MS/MS. (D) Images of mouse lung sections, stained using the Carstairs technique, highlighting the general histological changes across WT and Pla2g4a KO-infected mice with 500 TCID₅₀ of SARS-CoV-2/mouse (n = 4 to 8). (E) Assessment of lung inflammation scores among WT and Pla2g4a KO groups. (F) Heatmap of cytokine concentrations determined in lung homogenates through multiplex cytokines quantification (pg/mg of lung). Results are expressed as fold (log2) relative to WT-mock-infected mice (n = 8 to 10/group). Statistical differences were analyzed by Student’s t test or ordinary one-way ANOVA plus Fisher’s LSD test. Mann–Whitney test or Kruskal–Wallis plus Dunn’s posttest were carried out for nonparametric comparisons. ns: nonstatistically significant. (n = 5) *P < 0.05; **P < 0.005; ****P < 0.0001.

Fig. 2.

Modulation of the LMI signature by 12-LOX activity in the lungs of SARS-CoV-2-infected mice. (A) 12S-HETE, TXB2, 12-HEPE, and 12-HETrE dosages (pmol/mL) by LC–MS/MS in supernatants of mouse washed platelets stimulated with thrombin (1U/mL), or vehicle (Mock) for 15 min. (B) LMI biosynthetic pathways (C) 12(S)-HETE ELISA measurements (pg/mL) in plasma on Mock, 3 and 5 d postinfection. (D–Q) Individual graphs of the main LMIs (pmol/mg of lung tissue) that showed greater modulation with the loss of 12-LOX. (R) Dot plot with Log2 Fold change KO versus WT groups of the main LMI dosed over the course of infection. Lung LMI were analyzed by MetaboAnalyst 5.0. Statistical differences were analyzed by ordinary one-way ANOVA plus Fisher’s LSD test and Kruskal–Wallis test with Dunn’s posttest for nonparametric comparisons. ns: nonstatistically significant. (n = 5) *P < 0.05; **P < 0.005; ****P < 0.0001.

Fig. 3.

Lung and platelet transcriptomes during infection of WT and Alox12. RNA isolated from the lungs and platelets of mock and SARS-CoV-2-infected mice (n = 4 to 5) underwent RNA sequencing. (A) Principal component analysis of 27 lung samples based on normalized gene expression levels. (B–D) Gene Ontology enrichment analysis of molecular function, biological processes, and cellular component enriched in lung samples, comparing mice on day 3 and day 5 postinfection with mock-infected mice performed by Shiny GO 0.80 graphical gene-set enrichment tool. (E) Venn diagram illustrating the number of DEGs shared among or exclusive to each infected group compared to mock-infected mice. (F) Heat map representation of genes encoding the main inflammasome complex genes at 3 and 5 d postinfection normalized by the WT-mock-infected mice (n = 4 to 5). (G) Volcano plots illustrating differentially expressed genes (DEGs) in platelet samples of WT and Alox12 KO mock-infected mice. The numbers in the Left and Right sections of the volcano plots indicate the total downregulated and upregulated genes, respectively. (H) Gene Ontology pathway analysis (IPA software, QIAGEN) showing the most enhanced pathway.

Fig. 4.

12-LOX deficiency increases disease severity and lung inflammation after SARS-CoV-2 infection. Nine-wk-old WT and Alox12 KO mice were inoculated intranasally with 500 TCID₅₀/25 μL. (A) Changes in body weight during the infection course. The data are expressed as the mean ± SEM. From days 1 to 8 postinfection, differences were evaluated using a Two-way ANOVA followed by Fisher’s LSD test (n = 10 to 20). (B) Kaplan–Meier survival curve of WT and Alox12 KO-infected mice with 500 TCID₅₀ of SARS-CoV-2/mouse (n = 10 to 20). Lungs of mock- or SARS-CoV-2-infected mice were collected on days 3 and 5 postinfection. (C) The sum of the concentrations of key inflammatory cytokines was determined in lung homogenates through multiplex cytokine quantification (pg/mg of lung protein) in mock, 3, and 5 d postinfection groups (n = 10). (D) Whole blood male mice platelets activation determined by flow cytometry in mock, 3 and 5 d postinfection groups (n = 5 to 8). (E) Assessment of lung inflammation scores across different groups. (n = 8 to 9). (F) Mouse lung sections stained by Carstairs staining highlighting the general histological changes across the experimental groups. (G) Leukocytes density per lung area (mm²) among groups. Leukocytes were stained for CD3, Ly6G/Ly6C, and F4/80 markers and analyzed by confocal microscopy (n = 4 to 5). (H) Correlation heatmap displaying Spearman correlation coefficients between histopathological inflammatory scores and LMI dose (pmol/mg of lung protein). The color legend on the left side of the map illustrates Spearman correlation coefficients, while asterisks denote calculated P values. A P value < 0.05 was accepted as significant. Statistical differences were analyzed by ordinary one-way ANOVA plus Fisher’s LSD test. Nonparametric comparisons were conducted using the Kruskal–Wallis test with Dunn’s posttest. ns: nonstatistically significant. *P < 0.05; **P < 0.005; ****P < 0.0001.