Urokinase-type plasminogen activator and plasminogen activator inhibitor-1 complex as a serum biomarker for COVID-19

Abstract

Patients with coronavirus disease-2019 (COVID-19) have an increased risk of thrombosis and acute respiratory distress syndrome (ARDS). Thrombosis is often attributed to increases in plasminogen activator inhibitor-1 (PAI-1) and a shut-down of fibrinolysis (blood clot dissolution). Decreased urokinase-type plasminogen activator (uPA), a protease necessary for cell-associated plasmin generation, and increased tissue-type plasminogen activator (tPA) and PAI-1 levels have been reported in COVID-19 patients. Because these factors can occur in free and complexed forms with differences in their biological functions, we examined the predictive impact of uPA, tPA, and PAI-1 in their free forms and complexes as a biomarker for COVID-19 severity and the development of ARDS. In this retrospective study of 69 Japanese adults hospitalized with COVID-19 and 20 healthy donors, we found elevated free, non-complexed PAI-1 antigen, low circulating uPA, and uPA/PAI-1 but not tPA/PAI-1 complex levels to be associated with COVID-19 severity and ARDS development. This biomarker profile was typical for patients in the complicated phase. Lack of PAI-1 activity in circulation despite free, non-complexed PAI-1 protein and plasmin/α2anti-plasmin complex correlated with suPAR and sVCAM levels, markers indicating endothelial dysfunction. Furthermore, uPA/PAI-1 complex levels positively correlated with TNFα, a cytokine reported to trigger inflammatory cell death and tissue damage. Those levels also positively correlated with lymphopenia and the pro-inflammatory factors interleukin1β (IL1β), IL6, and C-reactive protein, markers associated with the anti-viral inflammatory response. These findings argue for using uPA and uPA/PAI-1 as novel biomarkers to detect patients at risk of developing severe COVID-19, including ARDS.

Keywords: C-reactive protein; COVID-19; fibrinolysis; interleukin-6; plasminogen activator inhibitor-1; respiratory distress syndrome; thrombosis; urokinase-type plasminogen activator.

Figure 1

Increased fibrinolysis with increases in plasmin/α2-antiplasmin in Japanese COVID-19 patients in the complicated phase. (A) Levels of D-dimers of COVID-19 patients in indicated clinical phases using ELISA. (B) Western blot analysis for plasminogen (B), plasmin/α2-antiplasmin complex (C), and α2-antiplasmin (D) on immunoblots after loading of equal amounts of protein in serum COVID-19 patient samples. All Western blots were performed at least twice with similar results. (B–D) Band intensity was quantified using signal intensity and normalized to one representative sample of a healthy donor. *p < 0.1; **p < 0.05; ***p < 0.01; n/s – not significant.

Figure 2

Free, non-complexed PAI-1 and uPA/PAI-1 complex levels in circulation are highest in severely ill COVID-19 patients. (A) Active PAI-1 (noncomplexed) levels were measured in patients and healthy donor serum using functional ELISA. Western blot analysis was used for serum level assessment of non-complexed PAI-1 antigen (B) with a molecular weight of 45 kDa, tPA (C) with a molecular weight of 70 kDa, tPA/PAI-1 complex with a molecular weight of 120 kDa (D), uPA (E, F) with a molecular weight of 50 kDa, and uPA/PAI-1 (E, G) complexes detectable as 100 Da band. (A–D, F–G) The quantification of band intensity of single proteins or protein complexes normalized to a sample from a healthy donor is given. (H, I) The Spearman coefficient r was used to correlate the interrelation of PAs/PAI-1 vs. plasmin/a2-antiplasmin (n = 36 per group). *p < 0.1; **p < 0.05; ***p < 0.01; n/s – not significant.

Figure 3

Alterations of endothelial dysfunction serum markers in COVID-19 patients related to plasmin activity. Serum soluble VCAM-1 (sVACM) (A) and uPAR (B) were measured using Western blot analysis, and the quantification of band intensity normalized to a sample from a healthy donor was given. The Spearman coefficient r was used for the correlation analysis of serum sVCAM1 and suPAR against PAI-1 activity (n = 45 per group) and non-complexed PAI-1 antigen (n = 36 per group) (C, D). Another Spearman correlation was done for sVCAM1 and suPAR against plasmin/α2AP) (Pm/α2AP) (n = 36 per group) (E, F). N depicts the number of patients tested in each group. *p < 0.1; **p < 0.05; ***p < 0.01; n/s – not significant.

Figure 4

Inflammatory markers and proinflammatory cytokines in COVID-19. ELISA measured CRP (A) and IL6 (B) antigen levels. Serum IL1β (C), TGFβ (D), and TNFα (E) were detected by Western blot analysis, and quantification of band intensity normalized to a sample from a healthy donor is given. (F) Fold change in TNFα transcript levels determined by qPCR. Expression was normalized using the β-actin expression in the same samples. Each sample was run in triplicate. N depicts the number of patients tested in each group. *p < 0.1; **p < 0.05; ***p < 0.01; n/s – not significant.

Figure 5

Correlation of fibrinolytic and inflammation markers in COVID-19 patients. (A) Heatmap of Spearman correlation coefficient. Pairs with positive coefficients are in brown, and those with negative coefficients are in blue. (B) P values associated with Spearman’s correlation of different parameters. (C) Heat map of Pearson’s correlation coefficient biomarkers of COVID-19 patients. The correlation coefficients are represented in the orange/blue color intensity change, shown in the color bar, whereby variables indicated by green or orange were positively or negatively correlated, respectively. Correlations were statistically significant (p<0.05) with 95% CIs for each correlation coefficient. (D) Model demonstrating the differences between COVID-19 patients in the uncomplicated (right side) and complicated (left side) phases. PAI-1, plasminogen activator inhibitor1); tPA, tissue-type plasminogen activator; uPA, urokinase plasminogen activator; SARS-CoV2, Severe acute respiratory syndrome coronavirus 2.