Identification of circulating microvesicle-encapsulated miR-223 as a potential novel biomarker for ARDS

Abstract

Acute respiratory distress syndrome (ARDS) is a lethal disease with severe forms conferring a mortality rate approaching 40%. The initial phase of ARDS results in acute lung injury (ALI) characterized by a severe inflammatory response and exudative alveolar flooding due to pulmonary capillary leak. Timely therapies to reduce ARDS mortality are limited by the lack of laboratory-guided diagnostic biomarkers for ARDS. The purpose of this study was to evaluate the prognostic role of circulating microvesicles (MVs)-containing miR-223 (MV-miR-223) if indicate more severe lung injury and worse outcomes in ARDS patients. Human plasma samples from one hundred ARDS patients enrolled in Albuterol to Treat Acute Lung Injury (ALTA) trial were compared to a control group of twenty normal human plasma specimens. The amount of MV-miR-223 was measured using absolute real-time polymerase chain reaction (PCR) with a standard curve. Mann-Whitney-Wilcoxon, Spearman correlation, Chi-squared tests, and Kaplan-Meier curves were computed to assess different variables and survival. Plasma levels of MV-miR-223 were significantly higher in ARDS patients compared to normal control subjects. Upon receiver operator characteristic (ROC) analysis of MV-miR-223 in relation to 30-day mortality, MV-miR-223 had an area under the curve (AUC) of 0.7021 with an optimal cut-off value of 2.413 pg/ml. Patients with high MV-miR-223 had higher 30-day mortality than subjects with low MV-miR-223 levels. MV-miR-223 was negatively correlated with ICU-free days, ventilator-free days, and organ failure-free days. Patients with high MV-miR-223 levels had higher 30 and 90-day mortality. MV-miR-223 was associated with 28-day clinical outcomes of ALTA trial including ICU-free days, ventilator-free days, and organ failure-free days. Thus, circulating MV-miR-223 may be a potential biomarker in prognosticating patient-centered outcomes and predicting mortality in ARDS.

Keywords: acute lung injury; extracellular vesicles; inflammation; microRNA; neutrophil; sepsis.

FIGURE 1

Purification and characterization of MVs derived from human plasma. (a) Schematic illustration of MV purification using sequential centrifugation protocols. (b) TEM images of plasma MVs (n = 4 per group) were shown, scale bar = 1 μm. (c) The diameters of isolated MVs (n = 3 per group) were measured using NTA. (d) MV positive markers (ANXA1, ANXA2, and ITGB1) and a negative marker (SP1) were detected in 100 μg pooled MV protein from the normal control group (n = 5), ARDS group (n = 5), and 20 μg THP‐1 cell lysate using western blot. ns, p > 0.05.

FIGURE 2

Comparison of plasma MV‐miR‐223 in different groups. (a) miR‐223 expression profile in various human primary cells from FANTOM5 database accessed on 05/18/2022. (b) normal control (n = 20) vs ARDS (n = 100). (c) Albuterol (n = 50) vs placebo (n = 50). (d) Direct lung injury (n = 54) vs indirect lung injury (n = 46) and (e) infectious (n = 66) vs non‐infectious (n = 28).

FIGURE 3

Plasma MV‐miR‐223 levels and mortality. (a) Survivors (n = 82) vs. non‐survivors (n = 18) at day 30. (b) MV‐miR‐223 distinguishes survivors from non‐survivors (area under the receiver operating characteristic curve [AUC] = 0.7021). A cut‐off of 2.413 pg/ml provided the highest AUC. (c) Kaplan–Meier survival curves for ARDS patients censored at 30‐days follow‐up. (d) Kaplan–Meier survival curves for ARDS patients censored at 90‐days follow‐up.

FIGURE 4

Correlation analyses between plasma MV‐miR‐223 and the prognostic parameters in ARDS patients. Spearman correlation coefficient was used to test relationships between (a) MV‐miR‐223 and APACHE III score (n = 98), (b) MV‐miR‐223 and PaO₂/FIO₂ ratio (n = 93), (c) MV‐miR‐223 and ICU‐free days (n = 100), (d) MV‐miR‐223 and ventilator‐free days (n = 100), (e) MV‐miR‐223 and organs failure‐free days (n = 98), (f) MV‐miR‐223 and bilirubin (n = 72). (g) MV‐miR‐223 and WBC counts (n = 38).

FIGURE 5

The copies of miR‐223 per MV are increased in ARDS patients. Plasma MVs were isolated from normal subjects and ARDS patients (n = 5 per group) using sequential centrifugation protocols. (a) The MV protein concentration is determined using BCA assay. (b) The number of MV particles is measured using NTA. (c) The copy number of miR‐223 in each MV was calculated. (d) miR‐142 expression profile in various human primary cells from FANTOM5 database accessed on 09/23/2022. (e) The relative levels of plasma MV‐miR‐142 in normal subjects and ARDS patients. (f) miR‐182 expression profile in various human primary cells from FANTOM5 database accessed on 09/23/2022. (g) The relative levels of plasma MV‐miR‐182 in normal subjects and ARDS patients. ns, p > 0.05.