Severe thermal and major traumatic injury results in elevated plasma concentrations of total heme that are associated with poor clinical outcomes and systemic immune suppression

Abstract

Background: Traumatic and thermal injuries result in a state of systemic immune suppression, yet the mechanisms that underlie its development are poorly understood. Released from injured muscle and lysed red blood cells, heme is a damage associated molecular pattern with potent immune modulatory properties. Here, we measured plasma concentrations of total heme in over 200 traumatic and thermally-injured patients in order to examine its relationship with clinical outcomes and post-injury immune suppression.

Methods: Blood samples were collected from 98 burns (≥15% total body surface area) and 147 traumatically-injured (injury severity score ≥8) patients across the ultra-early (≤1 hour) and acute (4-72 hours) post-injury settings. Pro-inflammatory cytokine production by lipopolysaccharide (LPS) challenged whole blood leukocytes was studied, and plasma concentrations of total heme, and its scavengers haptoglobin, hemopexin and albumin measured, alongside the expression of heme-oxygenase-1 (HO-1) in peripheral blood mononuclear cells (PBMCs). LPS-induced tumour necrosis factor-alpha (TNF-α) production by THP-1 cells and monocytes following in vitro heme treatment was also examined.

Results: Burns and traumatic injury resulted in significantly elevated plasma concentrations of heme, which coincided with reduced levels of hemopexin and albumin, and correlated positively with circulating levels of pro and anti-inflammatory cytokines. PBMCs isolated from trauma patients 4-12 and 48-72 hours post-injury exhibited increased HO-1 gene expression. Non-survivors of burn injury and patients who developed sepsis, presented on day 1 with significantly elevated heme levels, with a difference of 6.5 µM in heme concentrations corresponding to a relative 52% increase in the odds of post-burn mortality. On day 1 post-burn, heme levels were negatively associated with ex vivo LPS-induced TNF-α and interleukin-6 production by whole blood leukocytes. THP-1 cells and monocytes pre-treated with heme exhibited significantly reduced TNF-α production following LPS stimulation. This impairment was associated with decreased gene transcription, reduced activation of extracellular signal-regulated kinase 1/2 and an impaired glycolytic response.

Conclusions: Major injury results in elevated plasma concentrations of total heme that may contribute to the development of endotoxin tolerance and increase the risk of poor clinical outcomes. Restoration of the heme scavenging system could be a therapeutic approach by which to improve immune function post-injury.

Keywords: burns; critical care; heme; immune suppression; trauma.

Figure 1

Severe thermal injury results in an immediate and sustained systemic inflammatory response. (A–E) Comparison of concentrations of interleukin (IL)-6 (A), Granulocyte colony stimulator-factor (G-CSF) (B), Monocyte chemoattractant protein-1 (MCP-1) (C), IL-1 receptor antagonist (IL-1Ra) (D) and IL-10 (E) in serum samples obtained from thermally-injured patients at days 1 and 3 post-burn and healthy controls (HC). The number of samples analysed are stated below each study time-point. **p<0.005, ***p<0.0005.

Figure 2

Effect of severe thermal injury on lipopolysaccharide (LPS)-induced pro-inflammatory cytokine production by whole blood leukocytes. (A, B) Tumour necrosis factor-alpha (TNF-α; left panel) and interleukin-6 (IL-6; right panel) concentrations measured in supernatants of whole blood samples obtained from burns patients at days 1 and 3 post-injury and healthy controls (HC) following a 4 hour (A) or 18 hour (B) ex vivo stimulation with 10 ng/ml LPS. The number of samples analysed are stated below each study time-point. (C) Following a 24 hour culture in media supplemented with 10% serum obtained from burns patients on day 1 of injury or HC (n=5), THP-1 cells were stimulated with 1 µg/ml LPS, after which TNF-α concentrations in cell free supernatants (left panel) or TNF-α mRNA levels (right panel) were measured respectively. For the measurement of TNF-α concentrations in culture supernatants, THP-1 cells were stimulated for 4 hours with LPS. TNF-α mRNA levels were examined in THP-1 cells following a 2 hour stimulation with LPS. **p<0.005, ***p<0.0005 Vs vehicle. ^###p<0.0005.

Figure 3

Severe burns and major traumatic injury result in an immediate elevation in plasma concentrations of total heme. (A, B) Concentrations of total heme measured in PFP samples of severe burns (A) and major trauma (B) patients across the ultra-early (≤1H) and acute (4–72H) post-injury setting. The number of samples analysed are stated below each study time-point. **p<0.005, ***p<0.0005. HC, Healthy control.

Figure 4

Severe thermal injury but not major trauma results in the generation of fragmented red blood cells. (A, B) Absolute number (left panel) and frequency (right panel) of fragmented red blood cells (FRC) in peripheral blood samples of burns patients on days 1 and 3 post-injury (A) and major trauma patients (B) in the ultra-early (≤1H) and acute (4–72H) post-injury setting. The number of samples analysed are stated below each study time-point. *p<0.05. HC, Healthy control.

Figure 5

Effect of severe thermal injury and major trauma on the concentrations of plasma proteins implicated in heme and haemoglobin scavenging. (A‐C) Concentrations of hemopexin (A), albumin (B) and haptoglobin (C) measured in PFP samples acquired from thermally-injured patients on days 1 and 3 post-burn. (D‐F) Concentrations of hemopexin (D), albumin (E) and haptoglobin (F) measured in PFP samples acquired from major trauma patients during the ultra-early (≤1H) and acute (4–72H) post-injury phase. The number of samples analysed are stated below each study time-point. *p<0.05, **p<0.005, ***p<0.0005. HC, Healthy control.

Figure 6

Major traumatic injury but not severe burns results in increased heme oxygenase-1 (HO-1) gene expression in peripheral blood mononuclear cells (PBMCs). (A) Comparison of HO-1 mRNA levels in PBMCs of healthy controls (HC; n=10) and major trauma patients (n=20) at three post-injury time-points. (B) HO-1 gene expression in PBMCs isolated from HC (n=8) and severe burns patients (n=16) at days 1 and 3 post-thermal injury. **p<0.005.

Figure 7

Elevated plasma concentrations of total heme are associated with poor clinical outcomes in severe burns patients. (A) Total heme concentrations measured in PFP samples obtained on day 1 of injury from thermally-injured patients who did (n=42) or did not (n=37) develop sepsis post-burn. (B) Comparison of day 1 total heme levels in PFP samples of survivors (n=74) and non-survivors (n=14) of severe thermal injury. **p<0.005, ***p<0.0005.

Figure 8

In vitro heme treatment induces endotoxin tolerance in human monocytes. (A) Concentration of tumour necrosis factor-alpha (TNF-α) detected in supernatants of lipopolysaccharide (LPS) challenged THP-1 cells (n=6) pre-treated for 1 hour (left panel) or 4 hours (right panel) with 10 or 20 µM heme. **p<0.005, ***p<0.0005. (B) Concentration of TNF-α measured in supernatants of LPS challenged PBMCs (n=8) pre-treated for 4 hours with 10 or 20 µM heme. *p<0.05. (C) TNF-α levels recorded in supernatants of LPS challenged THP-1 cells pre-treated for 4 hours with 20 µM heme (n=8) or 100 ng/ml LPS (n=3) in media supplemented with 20% fetal calf serum (FCS). (D) Comparison of TNF-α mRNA levels in LPS stimulated THP-1 cells pre-treated for 4 hours with 20 µM heme or vehicle control (n=5). (E) Representative Western blot (top panel) and collated densitometry data (bottom panel, n=7) showing the phosphorylation status of the NF-κB subunit P65 in LPS challenged THP-1 cells pre-treated for 4 hours with 20 µM heme or vehicle control. **p<0.005 Vs. Baseline. B, Baseline; V, vehicle control; H, heme-treated.

Figure 9

Impaired lipopolysaccharide (LPS)-induced activation of the MAPK extracellular signal regulated kinase 1/2 (ERK 1/2) in heme pre-treated THP-1 cells. (A, B) Following a 1 hour pre-treatment with 10 µM PD98059 or vehicle control, THP-1 cells were stimulated with 1 µg/ml LPS, after which tumour necrosis factor-alpha (TNF-α) concentrations in culture supernatants (n=10) (A) and mRNA expression (n=6) (B) was measured. *p<0.05, ***p<0.0005. (C) Representative Western blot (left panel) and collated densitometry data (right panel, n=6) showing the phosphorylation status of ERK1/2 in LPS challenged THP-1 cells pre-treated for 4 hours with 20 µM heme or vehicle control. *p<0.05, ***p<0.0005 Vs Baseline. ^##p<0.005. B, Baseline; V, vehicle control; H, heme-treated.

Figure 10

Heme pre-treatment modulates lipopolysaccharide (LPS)-induced metabolic reprogramming in human monocytes. (A, B) Concentrations of lactate measured in supernatants of THP-1 cells (left panel, n=9) and PBMCs (right panel, n=11) challenged with 1 µg/ml LPS or vehicle control for 4 hours. (B) Concentration of tumour necrosis factor-alpha (TNF-α) detected in supernatants of lipopolysaccharide (LPS) challenged THP-1 cells (left panel) or PBMCs (right panel, n=11) pre-treated for 1 hour with 5–20 mM 2DG. For THP-1 cells, data represents n=21 (vehicle, 5 mM and 10 mM) and n=10 (20 mM). (C) Lactate concentrations in supernatants of LPS stimulated THP-1 cells (left panel, n=10) and PBMCs (right panel, n=11) pre-treated for 4 hours with 10–20 µM heme or vehicle control. *p<0.05, ***p<0.0005.