Traumatic Injury and Exposure to Mitochondrial-Derived Damage Associated Molecular Patterns Suppresses Neutrophil Extracellular Trap Formation

Abstract

Major traumatic injury induces significant remodeling of the circulating neutrophil pool and loss of bactericidal function. Although a well-described phenomenon, research to date has only analyzed blood samples acquired post-hospital admission, and the mechanisms that initiate compromised neutrophil function post-injury are therefore poorly understood. Here, we analyzed pre-hospital blood samples acquired from 62 adult trauma patients (mean age 44 years, range 19-95 years) within 1 h of injury (mean time to sample 39 min, range 13-59 min). We found an immediate impairment in neutrophil extracellular trap (NET) generation in response to phorbol 12-myristate 13-acetate (PMA) stimulation, which persisted into the acute post-injury phase (4-72 h). Reduced NET generation was accompanied by reduced reactive oxygen species production, impaired activation of mitogen-activated protein kinases, and a reduction in neutrophil glucose uptake and metabolism to lactate. Pre-treating neutrophils from healthy subjects with mitochondrial-derived damage-associated molecular patterns (mtDAMPs), whose circulating levels were significantly increased in our trauma patients, reduced NET generation. This mtDAMP-induced impairment in NET formation was associated with an N-formyl peptide mediated activation of AMP-activated protein kinase (AMPK), a negative regulator of aerobic glycolysis and NET formation. Indeed, activation of AMPK via treatment with the AMP-mimetic AICAR significantly reduced neutrophil lactate production in response to PMA stimulation, a phenomenon that we also observed for neutrophils pre-treated with mtDAMPs. Furthermore, the impairment in NET generation induced by mtDAMPs was partially ameliorated by pre-treating neutrophils with the AMPK inhibitor compound C. Taken together, our data demonstrate an immediate trauma-induced impairment in neutrophil anti-microbial function and identify mtDAMP release as a potential initiator of acute post-injury neutrophil dysfunction.

Keywords: immune suppression; mitochondrial-derived DAMPs; neutrophil extracellular traps; neutrophils; trauma.

Figure 1

Effect of traumatic injury on PMA-induced NET formation. (A) Basal NET generation by resting neutrophils isolated from healthy controls (HC) and trauma patients as assessed by DNA concentration in cell free supernatants following a 3 h in vitro culture. (B) HMGB-1 and (C) IL-33 concentrations in serum samples from HC and trauma patients. IL-33 levels were undetectable (N.D) in serum samples from HC. (D,E) Following a 3 h in vitro stimulation with PMA, NET production by neutrophils from HC and trauma patients was compared by measuring DNA concentration in cell free supernatants (D) and fluorescence microscopy (E). For supernatant analysis, number of samples are shown below each time-point. For microscope images, HC (n = 12), ≤1 h (n = 6), 4–12 h (n = 6), and 48–72 h (n = 8). ^*p < 0.01, ^**p < 0.001, ^***p < 0.0001 vs. HC.

Figure 2

Neutrophil ROS production and immature granulocyte count post-injury. Comparison of basal (A) and PMA-induced (B) ROS production by neutrophils isolated from healthy controls (HC) and trauma patients. Data are presented as area under the curve (AUC) (A) or fold increase above vehicle treated controls (B). (C,D) Frequency (C) and absolute number (D) of immature granulocytes (IG) in peripheral blood samples from healthy controls (HC) and trauma patients. Number of samples analyzed are shown below each time-point. ^*p < 0.01, ^**p < 0.001, ^***p < 0.0001 vs. HC.

Figure 3

Traumatic injury results in impaired PMA-induced activation of p38 MAPK. Phosphorylation status of P38 in resting and PMA-stimulated neutrophils isolated from healthy controls (HC) and trauma patients 4–12 and 48–72 h post-injury. Data are presented as representative Western blots (B) and densitometry analysis of p38 phosphorylation in resting (A) or PMA-stimulated neutrophils at the 4–12 h (C) and 48–72 h (D) post-injury time points. HC (n = 7), 4–12 h (n = 9), and 48–72 h (n = 4). For (A) ^*p < 0.01, ^**p < 0.001 vs. HC. For (C) and (D) ^**p < 0.001 vs. Time 0, ^#p < 0.01, ^###p < 0.0001 vs. HC sample at matched time point.

Figure 4

Traumatic injury results in impaired PMA-induced activation of ERK MAPK. Phosphorylation status of ERK 1/2 in resting and PMA-stimulated neutrophils isolated from healthy controls (HC) and trauma patients 4–12 and 48–72 h post-injury. Data are presented as representative Western blots (B) and densitometry analysis of ERK phosphorylation in resting (A) or PMA-stimulated neutrophils at the 4–12 h (C) and 48–72 h (D) post-injury time points. HC (n = 4), 4–12 h (n = 8), and 48–72 h (n = 6). For (C) and (D) ^*p < 0.01, ^**p < 0.001, ^***p < 0.0001 vs. Time 0, ^##p < 0.001, ^###p < 0.0001 vs. HC sample at matched time point.

Figure 5

Effect of traumatic injury on neutrophil glucose uptake and metabolism. Glucose uptake by basal (A) or PMA-stimulated (B) neutrophils isolated from healthy controls (HC) and trauma patients. MFI, Mean fluorescence intensity. (C,D) Comparison of lactate concentration in cell-free supernatants collected from resting (C) or PMA stimulated (D) neutrophils isolated from HC and trauma patients following a 3 h in vitro culture. Number of samples analyzed are indicated below each time-point. ^*p < 0.01, ^**p < 0.001, ^***p < 0.0001 vs. HC. (E,F) Expression of the glycolytic enzymes pyruvate kinase (PKM2) and lactate dehydrogenase A (LDHA) in resting neutrophils isolated from HC (n = 5) and trauma patients 48–72 h post-injury (n = 14). Data are presented as representative Western blots (E) and densitometry analysis of collated data for LDHA (F, top panel) and PKM2 (F, bottom panel). ^*p < 0.01 vs. HC.

Figure 6

Effect of mtDAMP pre-treatment on neutrophil NET generation. (A) Serum concentrations of the mitochondrial-derived N-formylated peptide ND6 in peripheral blood samples acquired from healthy controls (HC) and trauma patients. Number of samples are indicated below each time-point. ^***p < 0.0001 vs. HC. (B) NET production by PMA stimulated neutrophils pre-treated with 40 or 100 μg/ml mtDAMPs was assessed by measuring DNA content of cell-free supernatants (n = 5). ^*p < 0.01 vs. Vehicle. (C) MtDAMP-induced inhibition of NET generation was confirmed by fluorescence microscopy (n = 5; top panel x20 magnification; bottom panel x40 magnification). (D) ROS generation by neutrophils pre-treated with 100 μg/ml mtDAMPs or vehicle control was measured in response to stimulation with 25 nM PMA using luminol-based chemiluminescence (n = 10). Data are presented as area under the curve (AUC) and are mean ± SEM. ^***p < 0.0001 vs. PMA.

Figure 7

Treatment of neutrophils with mtDAMPs results in phosphorylation of AMPK. (A) Whole cell lysates prepared from purified neutrophils stimulated for 2–15 min with 100 μg/ml mtDAMPs were screened for phosphorylated AMPK. Western blot in top panel is representative of 4 independent experiments. For densitometry analysis ^***p < 0.0001 vs. 0 min. (B) AMPK phosphorylation in neutrophils treated for 1 h with the FPR-1 antagonist Cyclosporin H (CsH) or (C) or the CaMKK inhibitor STO-609 prior to a 5 min stimulation with 100 μg/ml mtDAMPs. Blots are representative of 5 (B) and 10 (C) independent experiments, with densitometric data depicted in the accompanying histogram. ^**p < 0.001, ^***p < 0.0001 vs. vehicle. (D) Comparison of PMA-induced NET formation by mtDAMP stimulated neutrophils pre-treated with the AMPK inhibitor compound C or vehicle control (n = 10). ^**p < 0.01 vs. PMA treated.

Figure 8

MtDAMP pre-treatment results in impaired neutrophil glycolysis. (A) Neutrophils pre-treated for 1 h with 1 mM AICAR or vehicle control were stimulated for 3 h with 25 nM PMA, after which lactate concentrations were measured in cell-free supernatants. Data are mean ± SEM of 12 independent experiments. ^***p < 0.0001. (B) Comparison of lactate concentrations in supernatants collected from neutrophils pre-treated with 100 μg/ml mtDAMPs or vehicle-control and subsequently stimulated with 25 nM PMA for 1, 2, and 3 h. Data are mean ± SEM of 10 independent experiments. ^**p < 0.001, ^***p < 0.0001 vs. vehicle.