Abstract

Abstract 3355

A significant number of injured patients with intermediate injury severity scores (15<ISS<30) develop multiple organ failure (MOF), which clinically begins with acute lung injury (ALI). Transfusion of >6 units of stored PRBCs (≥28 days) is associated with the development of ALI/MOF on day 3 post injury (Am J Surg 178:502-4, 1999). The pro-inflammatory mediators, e.g. cytokines, responsible for MOF/ALI in the injured have remained elusive; therefore, we hypothesize that “moonlighting” proteins, which have defined intracellular function when released in the circulation, activate innate immunity and are etiologic in the development of ALI/MOF post-injury.

Methods:

Proteomics on the field blood (plasma) of 3 patients with blunt trauma who later developed ALI/MOF and the plasma from 3 units of packed red blood cells (PRBCs) on day 1 and day 42 were completed using 2-dimensional gel electrophoresis/mass spectroscopy (MALDI/TOF) with computer analyses of the resultant peptides. The proteins from whole cell lysates from Human pulmonary microvascular endothelial cells (HMVECs) were separted by SDS-PAGE, transferred to nitrocellulse and immunoblotted with antibodies to protease activated receptor-1 (PAR-1) and PAR-2. HMVECs were also incubated for 6 hours and 1) ICAM-1 was measured by flow cytometry, 2) isolated neutrophils (PMNs) were added allowed to settle and in selected wells PMN adherence to these activated HMVECs was measured by myeloperoxidase content in the lysate, or 3) after the PMNs settled, lysophosphatidylcholines (lyso-PCs) [4.5μM], lipids from stored platelets implicated in TRALI, were added and the number of viable HMVECs/mm2 were counted by microscopy.

Results:

HMVECs display immunoreactivity for both PAR-1 and PAR-2. Of the 243 proteins identified in the injured patients and the stored vs. fresh PRBCs, α-enolase increased by 10.8-fold and 4.4-fold respectively (p<.05 & p<.005). Both thrombin and α-enolase induced ICAM-1 expression in HMVECs (Table 1) which was inhibited (60±8%) by pre-treatment with the anti-protease leupeptin. α-Enolase also induced significant PMN adhesion vs. media control: Media: 3.1±1.5; α-enolase (50 μg/ml): 14.4±4.7; LPS: 35.5±0.7*. The α-enolase-activated HMVECs vs. buffer-treated lyso-PCs induced significant PMN-mediated cytotoxicity (Table 2). We conclude that α-enolase from the injured and stored but not fresh PRBCs causes pro-inflammatory activation of HMVECs resulting in PMN adherence and PMN cytotoxicity in a two-event in vitro model through activation of PARs receptors. Moonlighting proteins like the glycolytic lyase α-enolase may have unexpected pro-inflammatory activity, which predispose the injured patient to increased morbidity and mortality.

Table 1:

ICAM-1 in HMVECs (MFI)

ControlThrombinα-Enolase
BufferLPS1 U/ml5 U/ml10 U/ml1 μg/ml10 μg/ml50 μg/ml
ICAM-1 (MFI) 687±178 1226±114* 1144±149 1528±213* 1362±211*    
1073±185 1881±234*    954±120 1311±186 2076±100* 
ControlThrombinα-Enolase
BufferLPS1 U/ml5 U/ml10 U/ml1 μg/ml10 μg/ml50 μg/ml
ICAM-1 (MFI) 687±178 1226±114* 1144±149 1528±213* 1362±211*    
1073±185 1881±234*    954±120 1311±186 2076±100* 
*

p<0.05 from Buffer Control

Table 2:

HMVEC Killing with Proteases (number of HMVECs)

ControlThrombin (5 U/ml)α-Enolase(50 μg/ml)
BufferLPS
Buffer 252.0±59.0 232.7±41.5 204.0±96.0 206.0±51.0 
LPC N/A N/A 207.0±109.0 215.3±55.1 
PMNs N/A 173.3±37.1 149.3±52.9 150.3±64.1 
LPC+PMNs N/A 88.7±22.6* 99.3±41.2*,  92.7±32.0*,  
ControlThrombin (5 U/ml)α-Enolase(50 μg/ml)
BufferLPS
Buffer 252.0±59.0 232.7±41.5 204.0±96.0 206.0±51.0 
LPC N/A N/A 207.0±109.0 215.3±55.1 
PMNs N/A 173.3±37.1 149.3±52.9 150.3±64.1 
LPC+PMNs N/A 88.7±22.6* 99.3±41.2*,  92.7±32.0*,  
*

p<0.05 from Buffer Control

p<0.05 from Protease Buffer Control

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.