Neutrophils are a main source of circulating suPAR predicting outcome in critical illness

BACKGROUND: Circulating levels of soluble urokinase plasminogen activation receptor (suPAR) have been proposed as a prognostic biomarker in patients with critical illness and sepsis. However, the origin of suPAR in sepsis has remained obscure. We investigated the potential cellular sources of suPAR by analyzing membrane-bound urokinase plasminogen activator receptor (uPAR, CD87) and evaluated its clinical relevance in critically ill patients. METHODS: We studied 87 critically ill patients (44 with sepsis, 43 without sepsis) from the medical intensive care unit (ICU) in comparison to 48 standard care patients with infections and 27 healthy controls in a prospective single-center non-interventional cohort study. Cellular uPAR expression of different immune cell subsets (by flow cytometry from peripheral blood) and corresponding serum suPAR concentrations were determined upon ICU admission and at day 3. Furthermore, we analyzed the effects of serum from sepsis patients on the activation and uPAR cleavage of primary human neutrophils and macrophages in vitro. RESULTS: In healthy controls, uPAR (CD87) expression was detected on nearly all blood neutrophils and monocytes, but only scarcely on lymphocytes. While uPAR expression on monocytes was maintained in ICU patients, only 58% of neutrophils from critically ill patients expressed uPAR, which was significantly lower than in healthy controls or standard care patients. Concomitantly, serum suPAR levels were significantly increased in ICU patients. We noted a clear inverse correlation between low neutrophilic uPAR and high serum suPAR in standard care and ICU patients, indicating that shedding of uPAR from activated neutrophils represents a main source of suPAR in systemic inflammation. Both low uPAR and high suPAR were closely associated with mortality in critically ill patients. Furthermore, serum from sepsis patients induced uPAR protein expression and subsequent receptor shedding on isolated primary neutrophils, but not on macrophages, in vitro. CONCLUSIONS: The inverse correlation between low uPAR surface expression on neutrophils and high serum suPAR in critically ill patients supports that neutrophils are a main source of shed suPAR proteins in systemic inflammation. Furthermore, high suPAR levels and low neutrophilic uPAR expression predict mortality in ICU patients.