- Email: [email protected]
Increased oxidative activity in human blood neutrophils and monocytes after spinal cord injury
Experimental Neurology 215 (2009) 308–316
Contents lists available at ScienceDirect
Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / yex n r
Increased oxidative activity in human blood neutrophils and monocytes after spinal cord injury Feng Bao a, Christopher S. Bailey b, Kevin R. Gurr b, Stewart I. Bailey b, M. Patricia Rosas-Arellano b, Gregory A. Dekaban a, Lynne C. Weaver a,⁎ a Spinal Cord Injury Team, BioTherapeutics Research Group, Robarts Research Institute, Schulich School of Medicine and Dentistry, University of Western Ontario, 100 Perth Drive, London, ON, Canada N6A 5K8 b Division of Orthopaedic Surgery, London Health Sciences Centre, University of Western Ontario, London, ON, Canada
a r t i c l e
i n f o
Article history: Received 29 July 2008 Revised 3 October 2008 Accepted 26 October 2008 Available online 13 November 2008 Keywords: Inflammation Oxidative damage Neutrophil Macrophage Oxidative enzyme Trauma
a b s t r a c t Traumatic injury can cause a systemic inflammatory response, increasing oxidative activity of circulating leukocytes and potentially exacerbating the original injury, as well as causing damage to initially unaffected organs. Although the importance of intraspinal inflammation after human spinal cord injury is appreciated, the role of the systemic inflammatory response to this injury is not widely recognised. We investigated oxidative activity of blood leukocytes from nine cord-injured subjects and six trauma controls (bone fractures without CNS injury) at 6 h–2 weeks after injury, comparing values to those of ten uninjured subjects. Neutrophil and monocyte free radical production, evaluated by flow cytometry, increased significantly more in cord injury subjects than in trauma controls (6-fold vs 50% increases). In leukocyte homogenates, the concentration of free radicals increased significantly more in cord injury subjects (2-fold) than in the trauma controls (1.6-fold) as did activity of myeloperoxidase (2.3-fold vs. 1.7-fold). Moreover, in homogenates and blood smears, expression of the NADPH oxidase subunit gp91phox and of the oxidative enzyme, inducible nitric oxide synthetase was 20–25% greater in cord injury subjects than in trauma controls. Expression of the pro-inflammatory transcription factor NF-κB and of cyclooxygenase-2 increased similarly after both injuries. Finally, aldehyde products of tissue-damaging lipid peroxidation also increased significantly more in the plasma of spinal cord injury subjects than in trauma controls (2.6 fold vs. 1.9-fold). Spinal cord injury causes a particularly intense systemic inflammatory response. Limiting this response briefly after cord injury should protect the spinal cord and tissues/organs outside the CNS from secondary damage. © 2008 Elsevier Inc. All rights reserved.
Introduction Primary spinal cord injury (SCI) is followed by secondary damage at the injury site, in which inflammation and a large intraspinal influx of leukocytes play an important role (Blight 1992; Fleming et al., 2006; Saville et al., 2004; Taoka and Okajima 1998; Tator and Fehlings 1991). In response to SCI, the release of leukocytes from marginal pool stores and the bone marrow into the circulation increases the number of cells available to cause damage (Furlan et al., 2006). The state of these leukocytes will impact on their actions upon entering the injured cord.
Abbreviations: ANOVA, analysis of variance; COX-2, cyclooxygenase-2; CNS, central nervous system; DHR123, dihydrorhodamine123; DCFH-DA, dichlorofluorescin diacetate; DCF, dichlorofluorescein; fMLP, formyl-methionyl-leucyl-phenylalanine; gp91phox, the catalytic subunit of NADPH oxidase; iNOS, inducible nitric oxide synthase; KPBS, potassium phosphate-buffered saline; MDA, malondialdehyde; MPO, myeloperoxidase; NADPH, nicotinamide adenine dinucleotide phosphate; R123, rhodamine123; RCF, relative centrifugal force; TBA, thiobarbituric acid; SCI, spinal cord injury; TBARS, thiobarbituric acid reactive substances. ⁎ Corresponding author. Fax: +1519663 3789. E-mail address: [email protected] (L.C. Weaver). 0014-4886/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2008.10.022
Leukocytosis is part of the well-known systemic inflammatory response to physical trauma; leukocyte activation consists of increased oxidative activity, phagocytic behaviour and migration (Bhatia et al., 2005; Shih et al., 1999; Tanaka et al., 1991). This inflammatory response can exacerbate the primary cord injury and also cause bystander damage to organs and tissues unaffected by the original injury (Baskaran et al., 2000; Bhatia et al., 2005; Utagawa et al., 2008). Animal studies have shown that CNS injury causes a more intense systemic inflammatory response than general trauma, with ensuing damage to organs such as the lungs and liver (Campbell et al., 2003; Gris et al., 2008). Although inflammation is a recognized source of secondary damage to the injured cord (Fleming et al., 2006), the intensity and character of a systemic inflammatory response to human SCI have not been studied. What are the destructive characteristics of activated circulating inflammatory cells? They may upregulate expression of oxidative enzymes, proteases and other potentially damaging molecules such as pro-inflammatory cytokines (Wang et al., 1997). Indeed, rat studies have confirmed that SCI causes a neutrophilia and changes the character of neutrophils, priming them for greater oxidative burst
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
activity (Gris et al., 2008). An increased capacity for oxidative burst in neutrophils, within hours of SCI in rats, can be induced by circulating pro-inflammatory cytokines and chemokines (Campbell et al., 2005; Wang et al., 1997), as well as by molecules such as platelet activating factor (Botha et al., 1996). Upon entry into the injured cord, these leukocytes would be primed to produce an oxidative burst, releasing proteolytic enzymes and inducing or extending the damage. Although the time-course and other characteristics of the inflammatory leukocyte infiltration into the injured human spinal cord have been studied (Fleming et al., 2006), the impact of SCI on these cells while still in the circulation is less well understood. Using flow cytometry, western blotting, biochemical analyses and immunohistochemical staining, we assessed the presence of reactive oxygen species, expression of a pro-inflammatory transcription factor, expression/activity of oxidative enzymes, and the concentration of products of cell membrane damage (lipid peroxidation) in blood samples obtained from SCI patients. Results were compared to those from a control group who suffered orthopaedic trauma with no CNS injury and from uninjured able-bodied subjects. Materials and methods Patient enrollment These studies were approved by the University of Western Ontario Research Ethics Board for the Review of Health Sciences Research Involving Human Subjects. Venous blood samples were obtained from twenty-five subjects after consent was obtained according to the Declaration of Helsinki. Exclusion criteria were polytrauma, a personal or family history of peripheral neuropathy or autoimmune disease, significant cognitive limitations, malignant cancer within five years of the study, chronic liver disease, regular medication with antiinflammatory drugs, diseases of the blood, pregnancy or lactation. None of the subjects were treated with steroids or other antiinflammatory drugs during the two-week period of the study.
309
The blood samples were taken, when possible, at 6, 12, 24, 48 and 72 h after the injury in trauma controls or spinal cord injury, as well as at 1 week and 2 week after the injuries. If a subject enrolled in the study required a blood transfusion, then no further samples were obtained from that patient. The SCI subjects were diagnosed by magnetic resonance imaging. The trauma controls had significant fractures of vertebrae or long bones without CNS injury. All SCI and trauma control subjects were thoroughly examined to exclude multisystem trauma and had computerized tomography scans of the thorax, abdomen and pelvis to exclude other associated injuries. The patients showed no evidence of systemic infection or local infection during the 2 week period of the study. All blood samples were taken under sterile condition and were analyzed in a laminar flow hood in a tissue culture laboratory. Details of these patients are in Table 1. Oxidative activity analysis Oxidative activity was detected by staining with dihydrorhodamine123 (DHR123), a non-fluorescent agent that is converted by cellular oxidation to the fluorescent dye rodamine123 (R123). Heparinized whole blood samples (200 μl) from uninjured, trauma control and SCI subjects were incubated with DHR123 (1.0 μM) in 800 μl RPMI Medium1640 at 37 °C in sterile conditions; the leukocytes were then isolated by ammonium chloride lysis of red blood cells. At least 50,000 events were analyzed for each blood sample. Neutrophils, monocytes and lymphocytes were initially gated by their characteristic forward and side scatter profiles, which represent size and granularity of the cells, respectively. Cells in these gates were then analyzed for fluorescence intensity. Cell-associated R123 fluorescence within the three populations was determined using a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA) and analyzed by FlowJo software (Tree Star, Ashland, OR). Background fluorescence (in samples incubated without R123) was subtracted from total fluorescence to obtain the value of oxidative activity from each individual sample.
Table 1 Description of subjects Subject Group
Spinal Level(s)
Case #
Age
Sex
Nature of accident
SCI
Trauma control
Uninjured
C5 T12 L1 T12 L1 C5-C7 T12 C6 L1 Fracture location T7-T10 C6-C7 Multiple Multiple Multiple Multiple
P#1 P#2 P#4 P#5 P#6 P#7 P#10 P#13 P#14
41 47 47 33 20 20 44 87 57
M F M M F M M M M
MVA Fall MVA MVA MVA MVA MVA Fall MVA
P#3 P#8 P#9 P#11 P#12 P#15 #1 #2 #3 #4 #5 #6 #7 #8 #9 #10
33 47 25 54 22 21 60 30 35 30 60 55 35 35 35 55
M M M F M F F M M F M M M M M M
MVA Fall MVA MVA MVA MVA
Initial assessmenta
Follow-up assessment
AIS
Motor Score
AIS
Motor score
Months post-SCI
A A A A C C A C A
10 50 52 50 60 40 50 53 60
–b B D A C D B D A
– 53 75 50 77 80 50 74 71
– 10 18 18 15 16 12 6 5
AIS, American Spinal Injury, Association Impairment Scale; C, Cervica; L, Lumbar; MVA, Motor Vehicle Accident; SCI, Spinal Cord Injury; T, Thoracic; a, At admission; b, Patient deceased at 60 h after SCI.
310
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
Myeloperoxidase and DCFH-DA assays Neutrophils express abundant MPO (Bao et al., 2004; Taoka et al., 1997) generating damaging hypochlorous acid. Macrophages also express MPO. Activity of the oxidative enzyme myeloperoxidase (MPO) was analyzed in leukocyte homogenates as described previously (Bao et al., 2004). Reactive oxygen species in the homogenates were detected by their oxidative conversion of the non-fluorescent 2′7′-dichlorofluorescein-diacetate (DCFH-DA) to the fluorescent 2′-7′dichlorofluorescein (DCF) (Bao et al., 2005). Immunocytochemical staining of blood smears A smear was made from blood of each sample and fixed in acetone overnight. The slides containing the smears were kept at −20°C until they were used. Blood smears were immunostained with antibodies to the oxidative enzymes inducible nitric oxide (iNOS, 1:2000, Calbiochem, San Diego, CA) and cyclooxygenase-2 (COX-2,1:500, Rockland, Gilbertsville, PA), to the transcription factor Nuclear Factor-κB p65 (NF-
κB, 1:1000, Abcam Inc, Cambridge MA), and to the catalytic subunit of nicotinamide adeninedinucleotide phosphate (NADPH) oxidase (gp91phox, 1:500, Upstate biotechnology, Lake Placid, KY) using methods described previously (Bao et al., 2004). NF-κB is activated by free radicals, inducing inflammatory gene transcription and cytokine secretion (Tak and Firestein 2001). gp91phox catalyzes the production of superoxide anions (Bao et al., 2004). All staining was performed using the same protocol. After quenching endogenous peroxidase activity, the smears were incubated overnight with the primary antibodies described above. They were then washed in PBS and incubated with biotinylated donkey anti-rabbit or donkey antimouse IgG (1:400) for 2 h. After rinsing in PBS, avidin-biotin complex (1:400) was applied for 2 h. The sections were then washed in PBS and immunoreactivity was visualized after incubation in a glucosediaminobenzidine-nickel solution for 5 min. The stained smears were rinsed in PBS, dehydrated through a gradient of ethanol, cleared, and coverslipped with Cytoseal mountant. PBS was substituted for the primary antibody on control smears in each reaction. A portion of the blood sample from four uninjured subjects was activated ex vivo with
Fig. 1. Oxidative burst in neutrophils and monocytes of uninjured subjects and of trauma controls or spinal cord injury subjects. (A) Example of gating of a blood sample for flow cytometry using physical characteristics of granularity (side scatter) and size (forward scatter). Neutrophils constituted the largest population whereas monocytes made up a smaller population. The lymphocyte population was not analyzed in this study. (B) Histograms plotting the number of cells vs. intensity of R123 fluorescence (log scale) produced by incubation of blood samples with DRH from uninjured subjects (U, orange lines) and from trauma controls (TC, red lines) and spinal cord injury (SCI, brown lines) subjects at 6 h, 24 h and 1 week after injury. Samples incubated without DHR revealed background fluorescence for uninjured (black line), trauma control (light green line) and SCI (blue line) subjects. Trauma control subjects had small increases in fluorescence due to oxidative burst in neutrophils and monocytes. Both neutrophils and monocytes had substantial oxidative activity at 24 h and 1 week after SCI.
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
the pro-inflammatory chemotactic peptide methionyl-leucyl-phenylalanine (fMLP (Gay et al., 1984), 2.5 nM) for 30 min at room temperature (24 °C) prior to preparing the smear to ascertain responses to a known inflammatory agent. For quantitative analysis of optical density of immunoreactive product, three smears from each subject (four subjects each group) were used and a total of ninety neutrophils was analyzed using Image Pro Plus software (Media Cybernetics, Bethesda, MD).
311
(ANOVA) and post-hoc Fisher's protected t-tests (Sokol and Rohlf 1981) or by a Student's t test when only two groups were compared. Sequential data collected during the first 72 h after SCI were compared to those of the uninjured subjects in a separate analysis from the values collected at 1 and 2 weeks after SCI, due to the smaller number of subjects available in the 1- and 2-week periods. Statistical significance was established at P ≤ 0.05. Results
Western blotting to detect iNOS and COX-2 and gp91phox expression Leukocyte counts and oxidative burst after SCI Standard methods for immunoblotting and analysis were used to detect iNOS, COX-2 and gp91phox, employing the antibodies listed above for immunocytochemistry (Bao et al., 2004). The leukocyte homogenates contained mostly neutrophils and lymphocytes, and a smaller number of monocytes, as confirmed by the flow cytometry studies above. As the oxidative activity of lymphocytes is small in comparison to that of neutrophils (Rothe, 1990), the findings in the assays of the leukocyte homogenates can be attributed mostly to neutrophils. Measurement of lipid peroxidation Lipid peroxidation products released from damaged tissues into the plasma were estimated by determining the relative levels of malondialdehyde (MDA), an aldehyde indicator of lipid peroxidation that was measured using a thiobarbiturate reactive substances (TBARS) assay (Bao et al., 2004). Statistical analysis Data are expressed as the mean ± standard error (SE). Differences among groups were established using one-way analysis of variance
The total leukocyte counts at 3.5 ± 1 h after injury in SCI (17.1 ± 1.4 × 109 cells/L) or trauma control (14.7 ± 1.2× 109 cells/L) subjects were greater than normal (4–10 × 109 cells/L). This increase was caused by a neutrophilia as neutrophil counts at this time after SCI were 14.8 ± 1.3 × 109 cells/L and in trauma controls they were 12.6 ± 0.9× 109 cells/L, compared to normal (2–7.5× 109 cells/L). In contrast, the monocyte counts in SCI and trauma controls (0.8± 0.1 × 109 and 0.7 ± 0.1 × 109 cells/L, respectively) were at the upper limit of normal (0.2–0.8× 109 cells/L). To assess oxidative burst, neutrophils, monocytes and lymphocytes were gated, using flow cytometry, by their characteristic forward and side scatter profiles (Fig. 1A) and then cells in each respective gate were analyzed for fluorescence intensity. Individual cases representative of the mean changes in the groups are shown in the histograms of Fig. 1B. After SCI the oxidative activity of neutrophils and monocytes was significantly increased by 4–6 fold in comparison to that of uninjured subjects at 12–48 h and 1 week post-injury (Figs. 2A and B; for details of statistics see Supplemental Table 1) and the maximal increases was at 24 h in neutrophil (6.4 fold). At 12 h, 24 h and 1w after SCI, the oxidative activity of neutrophils and monocytes of SCI subjects was significantly greater than that of trauma controls (Figs. 2A and B).
Fig. 2. Mean and individual changes in oxidative activity in neutrophils (A, C) and monocytes (B, D) after injury in trauma controls and spinal cord injury subjects. A, B. Oxidative activity is expressed as mean ± SE values of R123 fluorescence (after subtraction of background fluorescence) and samples are plotted at times encompassing 6–72 h and 1 and 2 weeks after injury. Uninjured subjects (U, open bars) had low oxidative activity. Oxidative burst in trauma control (TC, grey bars) subjects was slightly but insignificantly increased in circulating neutrophils (A), and monocytes (B). In contrast, oxidative activity increased significantly at 12, 24, 48 h and 1 week after spinal cord injury (SCI, black bars) in neutrophils (A) and monocytes (B). Increases after SCI were greater than those of trauma control subjects in neutrophils and monocytes at 12 h, 24 h and 1 week after injury. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t tests. #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t test. Statistical details in Supplemental Table 1. U: n = 7. At 6, 12, 24, 48, 72 h and 1 and 2 weeks after injury, TC: n = 5, 6, 6, 6, 5, 4 and 4; SCI: n = 4, 9, 9, 8, 5, 6 and 4. C, D. Mean fluorescence intensity (oxidative burst) in neutrophils and monocytes of seven individual uninjured subjects (open circles), three SCI subjects with cervical (C) injury (C5–7, open squares) and six subjects with high thoracic or lumbar injury (T–L, closed squares) are plotted as a scattergram. The responses did not differ according to segmental level of injury.
312
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
The oxidative activity of neutrophils and monocytes from trauma control subjects showed tendencies toward an increase that did not achieve statistical significance. However, when data from all seven sampling times were averaged, oxidative activity of neutrophils and monocytes of trauma control subjects was significantly greater than that of uninjured subjects by 52% and 55%, respectively (P = 0.047 and 0.025, see Supplemental Table 1). Lymphocytes had no evidence of oxidative burst activity when R123 fluorescence from SCI and trauma control subjects was compared to that of uninjured subjects (P = 0.421, data not shown). Of the nine SCI subjects, three had cervical injuries and six had injuries at either the 12th thoracic or 1st lumbar segment. We plotted the oxidative activity of the neutrophils and monocytes of the three subjects with the cervical injury separately from that of the six subjects with low thoracic or lumbar injuries to determine the distribution of response intensity with respect to injury location,. No segment-based pattern of intensity of oxidative burst was apparent in this scattergram (Figs. 2C and D). Particularly at the times (12–48 h) when the sample sizes were greatest, the responses within these subgroups of subjects greatly overlapped. Most responses fell outside the range of oxidative activity of the uninjured control subjects. Free radical formation in leukocytes
Fig. 3. Free radical production and MPO activity in leukocyte homogenates. (A) The presence of free radicals in the leukocytes was estimated by the conversion of DCFH to DCF in homogenates from uninjured (U) subjects and from trauma controls (TC) and spinal cord injury (SCI) subjects at times ranging from 6 h to 2 weeks after injury. DCF concentrations increased significantly in trauma controls and SCI subjects at most times assessed. Changes in leukocytes from SCI subjects were greater than those from trauma control subjects at 12 h, 24 h, 1 week and 2 weeks after injury. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t test. ##P ≤ 0.01; #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t test. U: n = 6. At 6, 12, 24, 48, 72 h and 1 and 2 weeks after injury, TC: n = 5, 6, 6, 6, 5, 4 and 4; SCI: n = 4, 7, 7, 7, 5, 6 and 4. (B) MPO activity increased at 6 h and 12 h after injury in trauma controls. In contrast, MPO activity increased significantly at all times assessed after SCI. The MPO activity at 24 h, 48 h, 72 h and 1 week after SCI was significantly greater than that in trauma control subjects. Definitions of ⁎⁎, ⁎, ##, # and number of subjects/group are as in (A). Statistical details in Supplemental Table 2.
The concentration of fluorescent DCF (the indicator of free radicals) was low in uninjured subjects (Fig. 3A). After injury, the DCF in leukocytes of both trauma controls and SCI subjects increased significantly from 6 h to 2 weeks after injury (P b 0.001, see Supplemental Table 2). Maximal changes were at 24 h after injury when DCF was increased by 1.6 fold in the trauma controls and by 2fold in the SCI subjects. The DCF concentrations were significantly greater in leukocytes of SCI subjects than in trauma controls at 12 h, 24 h and 1 week and 2 weeks after the injury. When data from all seven sampling times were averaged, DCF concentration due to free radical formation was significantly greater in SCI (by 20%) than in trauma control subjects (P = 0.006). Myeloperoxidase activity in leukocytes Samples from uninjured subjects exhibited low-level basal MPO activity (Fig. 3B). In the trauma controls MPO activity increased only at
Fig. 4. Western blot assay of (A) gp91phox, (B) iNOS and (C) COX-2 enzyme expression in leukocyte homogenates from uninjured (U), trauma controls (TC) and spinal cord injury (SCI) subjects at 24 h after injury. (A-C) Upper blots are typical western blots of enzyme expression in each group. Lower blots are β-actin expression demonstrating equal protein loading of gels for each enzyme. Bottom bar graphs show quantification of expression by densitometry, with enzyme expression expressed as a percent of β-actin density ± SE (n = 4 all groups). gp91phox, iNOS and COX-2 expression increased in trauma control and SCI subjects when compared to that of uninjured subjects. gp91phox and iNOS expression was greater after SCI than in trauma control subjects. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t test. ##P ≤ 0.01; #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t test. Statistical details in Supplemental Table 3.
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
6 h and 12 h after injury (P b 0.001, see Supplemental Table 2). In contrast, after SCI, MPO activity increased significantly from 6 h to 2 weeks after SCI. Maximal increases in MPO activity occurred at 6 h
313
after injury in trauma controls (1.7-fold increase) and at 24 h after SCI (2.3-fold increase). The MPO activity in the SCI subjects was significantly greater than that in trauma controls from 24 h to
Fig. 5. Expression of NF-κB, gp91Phox, iNOS and COX-2 in leukocytes of blood smears from uninjured (U), uninjured with fMLP activation (U+fMLP) and trauma controls (TC) and spinal cord injury (SCI) subjects at 24 h after injury. Blood smears were immunostained for the NF-κB (nucleus) and the oxidative enzymes (cytoplasm). Photomicrographs at the top of A-D are typical examples of immunostained cells. Intensely stained cells had morphology characteristic of neutrophils (see text). Calibration bars are 10 μm and apply to all panels. Ninety cells in total were analyzed by ImagePro software to obtain a measure of intensity of staining per cell (relative optical density/area). Bar graphs in lower panels of A–D depict mean relative optical density/area ± SE for each group (n = 4 subjects/group). Black and white arrow heads indicate the lobulated nucleus. After treatment of blood samples with fMLP and in trauma control and SCI subjects, expression of NF-κB, gp91Phox, iNOS and COX-2 all increased significantly when compared to values from uninjured subjects. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t test. ##P ≤ 0.01, significantly different from trauma controls by Fisher's protected t test, +P = 0.07, tended to differ from trauma controls by this test. Statistical details in Supplemental Table 4.
314
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
1 week after injury. Averaged activity from the seven sampling times also showed that MPO activity in SCI subjects was significantly greater (by 74%) than in trauma controls (P b 0.001). Expression of oxidative enzymes and NF-κB in leukocytes gp91phox protein expression was readily detected in leukocytes of uninjured subjects (Fig. 4A). At 24 h after injury in trauma control or SCI subjects, this gp91phox expression increased significantly (1.5-fold and 1.8-fold, respectively) compared to the uninjured subjects and expression was significantly greater in SCI subjects than in trauma controls (P b 0.001, see Supplemental Table 3). Expression of iNOS increased from the low levels in the uninjured subjects to significantly greater expression at 24 h after either type of injury (Fig. 4B, P b 0.001). Again, increases after SCI (2.2-fold) were greater than those in trauma controls (1.8-fold). Finally, expression of COX-2 also increased significantly in leukocytes of trauma control (4.2-fold) or SCI (4.9fold) subjects when compared to the lower levels in uninjured subjects (Fig. 4C, P b 0.001). The robust increases in COX-2 expression after either type of injury were not significantly different from each other. Immunocytochemical staining of leukocytes in blood smears confirmed expression of these oxidative enzymes and of the proinflammatory transcription factor NF-κB at 24 h after injury and revealed the phenotype of cells expressing these molecules. Expression was consistently found in neutrophils identified by their size (diameter 10–12 μm) and lobular nucleus. Cells with morphology of monocytes or lymphocytes exhibited low levels of expression of each of these proteins (data not shown). Little expression of NF-κB was detected in neutrophils from uninjured subjects (Fig. 5A top panels). In contrast, expression of NF-κB was clear after incubation of samples from uninjured subjects after fMLP stimulation and at 24 h after injury in trauma controls and SCI subjects. Active NF-κB was mostly in the nucleus with fainter expression in the cytoplasm (Fig. 5A top panels). Quantification of expression by densitometry revealed significant, approximately 2-fold increases after treatment with fMLP, as well as in samples from trauma controls and SCI subjects, when compared to that in the uninjured subjects (Fig. 5A lower panel, P b 0.001, see Supplemental Table 4). Expression of gp91phox, iNOS and COX-2 in neutrophils was confined to the cytoplasm and appeared at low levels even in leukocytes of uninjured subjects (Figs. 5B–D top panels). Treatment of blood samples from uninjured subjects with fMLP significantly increased expression of gp91phox, iNOS and COX-2 by 1.6-fold, 1.6-fold and 2fold, respectively (P b 0.001). In trauma controls, expression of these three enzymes increased significantly by 1.5-fold, 1.8-fold and 2-fold, respectively, when compared to that in uninjured subjects (Figs. 5B–D, lower panels). Likewise, after SCI, expression of gp91phox, iNOS and COX-2 increased by 1.8-fold, 2.7-fold and 2.3-fold, respectively. Increases in iNOS expression in leukocytes of SCI subjects were significantly greater than those in trauma controls. The tendency for increased gp91phox expression in SCI subjects compared to trauma controls almost attained statistical significance (P = 0.07). Lipid peroxidation The TBARS concentration (revealing lipid peroxidation products) was low in the plasma of uninjured subjects (Fig. 6). In trauma controls, the plasma TBARS concentration increased significantly above that in uninjured subjects at 24–72 h and at 2 weeks after the injury (P = 0.022, 6–72 h; P = 0.018, 1, 2 weeks, see Supplemental Table 5). The maximal (1.9-fold) increase occurred at 24 h after injury. In contrast, the plasma TBARS concentration increased significantly from 6 h to 2 weeks after SCI, when compared to values in uninjured subjects. Again maximal increases (2.6-fold) occurred at 24 h after injury. At 12–48 h and 1 week after injury, the plasma TBARS concentration in SCI subjects was
Fig. 6. Mean changes in products of lipid peroxidation detected in plasma samples obtained from uninjured (U), trauma controls (TC) and spinal cord injury (SCI) subjects. The products were primarily malondialdehye, detected by the thiobarbituric acid reactive substances (TBARS) assay, and expressed as TBARS concentration/mg protein (mean ± SE) at times from 6–72 h and 1 and 2 weeks after injury. Lower levels of TBARS were present in the serum of uninjured subjects and this concentration increased significantly at 24–72 h and 2 weeks in trauma controls. In contrast, TBARS concentrations increased significantly at all times measured after SCI and were significantly higher than those in trauma controls at 12–48 h and 1 week after injury. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t tests. ##P ≤ 0.01; #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t tests. Statistical details in Supplemental Table 5. U: n = 6. At 6, 12, 24, 48, 72 h and 1 and 2 weeks after injury, TC: n = 5, 6, 6, 6, 5, 4 and 4; SCI: n = 4, 7, 7, 7, 5, 6 and 4.
significantly greater than in trauma controls (Fig. 6). When data from all seven sampling times were averaged, the plasma concentration of TBARS in SCI subjects was significantly greater than that in trauma controls (P b 0.0001). This measure of lipid peroxidation in the subjects with cervical injury (C5–7, n = 3) at 24 h and 48 h after injury (335 ± 11 and 293 ± 11 nmol/100 mg protein) was not different from that in subjects with low thoracic or lumbar injury (T12-L1, n = 4) SCI (359 ± 75 and 385 ± 107 nmol/100 mg protein). Discussion Following traumatic SCI, neutrophils and then monocyte/macrophages infiltrate the lesion from the circulation (Fleming et al., 2006). Our study reveals that insult to the spinal cord caused a particularly intense induction of oxidative burst and associated enzymes in the circulating leukocytes, significantly greater than that in the trauma control subjects. The highly activated neutrophils and monocytes after SCI could play a significant role in the secondary damage that occurs within and outside the injured spinal cord. The different parameters measured in our study correlated well with each other. Increased oxidative burst synchronized with increased free radical production, MPO activity and expression of the key enzymes in the induction of oxidative burst (gp91phox, iNOS and COX-2). These changes paralleled evidence of secondary tissue damage due to lipid peroxidation. Expression of the pro-inflammatory transcription factor NF-κB also increased. As neutrophils are the first leukocytes to enter the injured human spinal cord (Fleming et al., 2006), their robust oxidative activity likely contributes to the early stages of secondary damage to surviving tissue. We used several different methods to detect changes in leukocyte oxidative activity after SCI. First, using flow cytometry we found that the neutrophils and monocytes increased their oxidative activity intensely in the first two days after the injury. Lymphocytes did not undergo oxidative burst activity. At the peak time of oxidative activity (24 h after SCI), we extended the investigation to examine oxidative enzymes and free radical production using western blot and biochemical analysis from leukocyte homogenates. As monocytes constitute only a small portion of the white cell population, our data from the leukocyte homogenates likely reflect the responses of neutrophils. Similarly, few monocytes were detected in the blood
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
smears. The increased oxidative activity may well have been present in the monocytes but, with the exception of the flow cytometry experiments, we did not isolate this small population and they likely made a small contribution to the responses that we measured. Studies in rats and mice (Blight 1992; Saville et al., 2004) have shown that neutrophils generate reactive oxygen and nitrosyl radicals as well as a variety of damaging oxidative enzymes (Bao et al., 2004; Bao et al., 2005). Neutrophils also produce pro-inflammatory cytokines and chemokines (Brandt et al., 2000; Cassatella et al., 1995). Cytokine and chemokine gene expression in neutrophils depends on activation of the pro-inflammatory transcription factor NF-kB (McDonald et al., 1997) and the NF-κB activation in circulating neutrophils in our study suggests that these cells were producing proinflammatory and oxidative proteins. Ours is the first demonstration of activated NF-kB in circulating neutrophils after traumatic SCI. Induction of hepatic NF-kB recently has been shown to play a key role in activation and recruitment of neutrophils into the injured mouse brain (Campbell et al., 2008). A robust systemic inflammatory response to severe trauma or infection often culminates in multiple organ failure (Moore et al., 1994). The multiple organ dysfunction after severe general trauma in humans, such as that caused by multiple bone fractures or extensive burns, is mediated, in part, by neutrophils that become activated and more responsive to stimulation (Bhatia et al., 2005). For example, neutrophils infiltrate the lungs in conditions that produce a systemic inflammatory response such as burns, multiple trauma, and infections in humans and/or animal models (Bhatia et al., 2005; Gris et al., 2008; Sunil et al., 2002). A recent study in rats revealed that a systemic inflammatory response to SCI can cause a major influx of neutrophils into the lungs, leading to oxidative and proteolytic enzyme expression/activity within the organ (Gris et al., 2008). This results in lung damage, revealed by the presence of lipid peroxidation. Brain and spinal cord injury also causes an influx of leukocytes into the liver and causes liver damage, due to the almost immediate triggering of hepatic chemokine and acute phase protein expression and release (Campbell et al., 2005; Wilcockson et al., 2002). Such systemic inflammation likely contributes to organ dysfunction after SCI, as it does after brain injury and other traumatic insults (Baskaran et al., 2000; Bhatia et al., 2005; Ott et al., 1994). Therefore, the activation of leukocytes by SCI, and downstream consequences of this activation, not only exacerbates the spinal injury, but also puts key organs such as the lungs and liver at risk. Indeed respiratory failure is the leading cause of short-term and long-term mortality after human SCI (DeVivo et al., 1999) and, in our small sample of subjects, one died of respiratory failure and a second experienced severe pulmonary complications. The most important finding in our study was that the systemic inflammatory response after SCI was usually greater than the response to severe bone fractures. The difference was not in the degree of soft tissue or bone damage but instead appears to correlate with injury to the CNS. The mechanism for this intense reaction is unknown but brain injury also causes a particularly intense systemic inflammatory response (Campbell et al., 2005). After direct CNS injury, loss of crucial feedback control of immune function occurs. For example, significant cord injury at all levels releases opiates (Bakshi et al., 1990) that can cause neutrophil activation (Menzebach et al., 2003), contributing to the systemic inflammatory response. In addition, injury to the upper thoracic or cervical cord severely depresses sympathetic activity and catecholamine levels (Maiorov et al., 1997; Mathias et al., 1975). As catecholamines inhibit neutrophil oxidative burst and suppress expression of integrins (Trabold et al., 2007), the low catcholamine levels after SCI also are proinflammatory. These effects of CNS injury on immune function, would not occur after general trauma. The systemic inflammatory response that we report in the SCI and trauma control subjects is a well-known response to trauma. However, a significant literature also reports immunosuppression after trauma,
315
including SCI (Campagnolo et al., 1997; Cruse et al., 2000; Meisel et al., 2005). This immunosuppression primarily affects the function of natural killer cells, T-cells, B cells and macrophages in humans and animals at days to weeks after injury (Cruse et al., 1992; Cruse et al., 2000; Furlan et al., 2006; Lucin et al., 2007; Riegger et al., 2007). Unlike the responses in our study, the depressant effects of SCI on immune function are mostly time-dependent with greater suppressive effects in the chronic phases of injury. Immunosuppression also can very with the level of SCI, with cervical or high thoracic injuries having greater effects on lymphocytes and phagocytic properties of neutrophils (Campagnolo et al., 1997; Lucin et al., 2007; Meisel et al., 2005). This response is independent of time after SCI. In contrast, the intensity of oxidative burst and lipid peroxidation in our study did not vary with the segment of injury. In conclusion, SCI causes a profound systemic inflammatory response that may have widespread consequences. This response within the circulation is a ready target for intravenous intervention with agents that transiently limit the migration of neutrophils and monocytes into the injured spinal cord or visceral organs. Acknowledgments We especially thank all of the volunteers and patients for contributing to this study. We appreciate that the SCI and trauma subjects made the decision to contribute to this research, and permitted the blood samples to be drawn, under circumstances when their own personal needs were very great. We also thank all of the fellows, nurses and staff at the University Health Sciences Centre Victoria Hospital Campus and Robarts Research Institute who facilitated our work by coordinating the timely acquisition and transport of samples to our laboratory. We appreciate the advice of our flow cytometry manager, Dr. Kristin Chadwick, assistance with the assays by summer student Ms. Maria Pineda, and data preparation by Ms. Eilis Hamilton. The authors appreciate the constructive critique of the manuscript by Dr. Canio Polosa. This research was funded by a New Emerging Team grant from the Canadian Institutes of Health Research (RMN63187). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.expneurol.2008.10.022. References Bakshi, R., Newman, A.H., Faden, A.I., 1990. Dynorphin A-(1–17) induces alterations in free fatty acids, excitatory amino acids, and motor function through an opiatereceptor-mediated mechanism. J. Neurosci. 10 (12), 3793–3800. Bao, F., Chen, Y., Dekaban, G.A., Weaver, L.C., 2004. Early anti-inflammatory treatment reduces lipid peroxidation and protein nitration after spinal cord injury in rats. J. Neurochem. 88, 1335–1344. Bao, F., Dekaban, G.A., Weaver, L.C., 2005. Anti-CD11d antibody treatment reduces free radical formation and cell death in the injured spinal cord of rats. J. Neurochem. 94, 1361–1373. Baskaran, H., Yarmush, M.L., Berthiaume, F., 2000. Dynamics of tissue neutrophil sequestration after cutaneous burns in rats. J. Surg. Res. 93, 88–96. Bhatia, R.K., Pallister, I., Dent, C., Jones, S.A., Topley, N., 2005. Enhanced neutrophil migratory activity following major blunt trauma. Injury 36, 956–962. Blight, A.R., 1992. Macrophages and inflammatory damage in spinal cord injury. J. Neurotrauma 9 (Supp 1), S83–S91. Botha, A.J., Moore, F.A., Moore, E.E., Peterson, V.M., Silliman, C.C., Goode, A.W., 1996. Sequential systemic platelet-activating factor and interleukin 8 primes neutrophils in patients with trauma at risk of mulitple organ failure. Br. J. Surg. 83, 1407–1412. Brandt, E., Woerly, G., Younes, A.B., Loiseau, S., Capron, M., 2000. IL-4 production by human polymorphonuclear neutrophils. J. Leukocyte Biol. 68, 125–130. Campagnolo, D.I., Bartlett, J.A., Keller, S.E., Sanchez, W., Oza, R., 1997. Impaired phagocytosis of Staphylococcus aureus in complete tetraplegics. Am. J. Phys. Med. Rehabil. 76 (4), 276–280. Campbell, S.J., Hughes, P.M., Iredale, J.P., Wilcockson, D.C., Waters, S., Docagne, F., Perry, V.H., Anthony, D.C., 2003. CINC-1 is an acute-phase protein induced by focal brain injury causing leukocyte mobilization and liver injury. FASEB J. 17, 1168–1170.
316
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
Campbell, S.J., Perry, V.H., Pitossi, F.J., Butchart, A.G., Chertoff, M., Waters, S., Dempster, R., Anthony, D.C., 2005. Central nervous system injury triggers hepatic CC and CXC chemokine expression that is associated with leukocyte mobilization and recruitment to both the central nervous system and the liver. Am. J. Pathol. 166, 1487–1497. Campbell, S.J., Anthony, D.C., Oakley, F., Carlsen, H., Elsharkawy, A.M., Blomhoff, R., Mann, D.A., 2008. Hepatic nuclear factor kappa B regulates neutrophil recruitment to the injured brain. J. Neuropathol. Exp. Neurol. 67 (3), 223–230. Cassatella, M.A., Gasperini, S., D'Andrea, A., Ma, X., Trinchieri, G., Meda, L., 1995. Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25, 1–5. Cruse, J.M., Lewis, R.E., Bishop, G.R., Kliesch, W.F., Gaitan, E., 1992. Neuroendocrineimmune interactions associated with loss and restoration of immune system function in spinal cord injury and stroke patients. Immunol. Res. 11 (2), 104–116. Cruse, J.M., Lewis, R.E., Dilioglou, S., Roe, D.L., Wallace, W.F., Chen, R.S., 2000. Review of immune function, healing of pressure ulcers, and nutritional status in patients with spinal cord injury. J. Spinal Cord Med. 23 (2), 129–135. DeVivo, M.J., Krause, J.S., Lammertse, D.P., 1999. Recent trends in mortality and causes of death among persons with spinal cord injury. Arch. Phys. Med. Rehabil. 80 (11), 1411–1419. Fleming, J.C., Norenberg, M.D., Ramsay, D.A., Dekaban, G.A., Marcillo, A.E., Saenz, A.D., Pasquales-Style, M., Dietrich, W.D., Weaver, L.C., 2006. The cellular inflammatory response in human spinal cords after injury. Brain 129 (Pt 12), 3249–3269. Furlan, J.C., Krassioukov, A.V., Fehlings, M.G., 2006. Hematologic abnormalities within the first week after acute isolated traumatic cervical spinal cord injury: a casecontrol cohort study. Spine 31, 2674–2683. Gay, J.C., Beckman, J.K., Brash, A.R., Oates, J.A., Lukens, J.N., 1984. Enhancement of chemotactic factor-stimulated neutrophil oxidative metabolism by leukotriene B4. Blood 64, 780–785. Gris, D., Hamilton, E.F., Weaver, L.C., 2008. The systemic inflammatory response after spinal cord injury damages lungs and kidneys. Exp. Neurol. 211, 259–270. Lucin, K.M., Sanders, V.M., Jones, T.B., Malarkey, W.B., Popovich, P.G., 2007. Impaired antibody synthesis after spinal cord injury is level dependent and is due to sympathetic nervous system dysregulation. Exp. Neurol. 207 (1), 75–84. Maiorov, D.N., Weaver, L.C., Krassioukov, A.V., 1997. Relationship between sympathetic activity and arterial pressure in conscious spinal rats. Am. J. Physiol. 272, H625–H631. Mathias, C.J., Christensen, N.J., Corbett, J.L., Frankel, H.L., Goodwin, T.J., Peart, W.S., 1975. Plasma catecholamines, plasma renin activity and plasma aldosterone in tetraplegic man, horizontal and tilted. Clin. Sci. Mol. Med. 49, 291–299. McDonald, P.P., Bald, A., Cassatella, M.A., 1997. Activation of the NF-kappaB pathway by inflammatory stimuli in human neutrophils. Blood 89, 3421–3433. Meisel, C., Schwab, J.M., Prass, K., Meisel, A., Dirnagl, U., 2005. Central nervous system injury-induced immune deficiency syndrome. Nat. Rev. Neurosci. 6 (10), 775–786.
Menzebach, A., Hirsch, J., Hempelmann, G., Welters, I.D., 2003. Effects of endogenous and synthetic opioid peptides on neutrophil function in vitro. Br. J. Anaesth. 91 (4), 546–550. Moore, E.E., Moore, F.A., Franciose, R.J., Kim, F.J., Biffl, W.L., Banerjee, A., 1994. The postischemic gut serves as a priming bed for circulating neutrophils that provoke multiple organ failure. J. Trauma 37 (6), 881–887. Ott, L., McClain, C.J., Gillespie, M., Young, B., 1994. Cytokines and metabolic dysfunction after severe head injury. J. Neurotrauma 11, 447–472. Riegger, T., Conrad, S., Liu, K., Schluesener, H.J., Adibzahdeh, M., Schwab, J.M., 2007. Spinal cord injury-induced immune depression syndrome (SCI-IDS). Eur. J. Neurosci. 25 (6), 1743–1747. Rothe, G.V.G., 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidin and 2´,7´-dichlorofluorescin. J. Leukocyte Biol. 47, 440–448. Saville, L.R., Pospisil, C.H., Mawhinney, L.A., Bao, F., Simedria, F.C., Peters, A.A., O'Connell, P.J., Weaver, L.C., Dekaban, G.A., 2004. A monoclonal antibody to CD11d reduces the inflammatory infiltrate into the injured spinal cord: A potential neuroprotective treatment. J. Neuroimmunol. 156, 42–57. Shih, H.C., Su, C.H., Lee, C.H., 1999. Superoxide production of neutrophils after severe injury: impact of subsequent surgery and sepsis. Am. J. Emerg. Med. 17 (1), 15–18. Sokol, R.R., Rohlf, F.J., 1981. Biometry: The Principles and Practice of Statistics in Biological Research. W.H. Freeman, San Francisco. Sunil, V.R., Connor, A.J., Zhou, P., Gordon, M.K., Laskin, J.D., Laskin, D.L., 2002. Activation of adherent vascular neutrophils in the lung during acute endotoxemia. Respir. Res. 3 (1), 21. Tak, P.P., Firestein, G.S., 2001. NFkB: a key role in inflammatory diseases. J. Clin. Invest. 107, 7–11. Tanaka, H., Ogura, H., Yokota, J., Sugimoto, H., Yoshioka, T., Sugimoto, T., 1991. Acceleration of superoxide production from leukocytes in trauma patients. Ann. Surg. 214 (2), 187–192. Taoka, Y., Okajima, K., 1998. Spinal cord injury in the rat. Prog. Neurobiol. 56, 341–358. Taoka, Y., Okajima, K., Uchiba, M., Murakami, K., Kushimoto, S., Johno, M., Naruo, M., Okabe, H., Takatsuki, K., 1997. Role of neutrophils in spinal cord injury in the rat. Neuroscience 79 (4), 1177–1182. Tator, C.H., Fehlings, M.G., 1991. Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms. J. Neurosurg. 75, 15–26. Trabold, B., Gruber, M., Frohlich, D., 2007. Functional and phenotypic changes in polymorphonuclear neutrophils induced by catecholamines. Scand. Cardiovasc. J. 41 (1), 59–64. Utagawa, A., Truettner, J.S., Dietrich, W.D., Bramlett, H.M., 2008. Systemic inflammation exacerbates behavioral and histopathological consequences of isolated traumatic brain injury in rats. Exp. Neurol. 211 (1), 283–291. Wang, C.X., Olschowka, J., Wrathall, J.R., 1997. Increase of interleukin-1beta mRNA and protein in the spinal cord following experimental traumatic injury in the rat. Brain Res. 759, 190–196. Wilcockson, D.C., Campbell, S.J., Anthony, D.C., Perry, V.H., 2002. The systemic and local acute phase response following acute brain injury. J. Cereb. Blood Flow Metab. 22, 318–326.
Contents lists available at ScienceDirect
Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / yex n r
Increased oxidative activity in human blood neutrophils and monocytes after spinal cord injury Feng Bao a, Christopher S. Bailey b, Kevin R. Gurr b, Stewart I. Bailey b, M. Patricia Rosas-Arellano b, Gregory A. Dekaban a, Lynne C. Weaver a,⁎ a Spinal Cord Injury Team, BioTherapeutics Research Group, Robarts Research Institute, Schulich School of Medicine and Dentistry, University of Western Ontario, 100 Perth Drive, London, ON, Canada N6A 5K8 b Division of Orthopaedic Surgery, London Health Sciences Centre, University of Western Ontario, London, ON, Canada
a r t i c l e
i n f o
Article history: Received 29 July 2008 Revised 3 October 2008 Accepted 26 October 2008 Available online 13 November 2008 Keywords: Inflammation Oxidative damage Neutrophil Macrophage Oxidative enzyme Trauma
a b s t r a c t Traumatic injury can cause a systemic inflammatory response, increasing oxidative activity of circulating leukocytes and potentially exacerbating the original injury, as well as causing damage to initially unaffected organs. Although the importance of intraspinal inflammation after human spinal cord injury is appreciated, the role of the systemic inflammatory response to this injury is not widely recognised. We investigated oxidative activity of blood leukocytes from nine cord-injured subjects and six trauma controls (bone fractures without CNS injury) at 6 h–2 weeks after injury, comparing values to those of ten uninjured subjects. Neutrophil and monocyte free radical production, evaluated by flow cytometry, increased significantly more in cord injury subjects than in trauma controls (6-fold vs 50% increases). In leukocyte homogenates, the concentration of free radicals increased significantly more in cord injury subjects (2-fold) than in the trauma controls (1.6-fold) as did activity of myeloperoxidase (2.3-fold vs. 1.7-fold). Moreover, in homogenates and blood smears, expression of the NADPH oxidase subunit gp91phox and of the oxidative enzyme, inducible nitric oxide synthetase was 20–25% greater in cord injury subjects than in trauma controls. Expression of the pro-inflammatory transcription factor NF-κB and of cyclooxygenase-2 increased similarly after both injuries. Finally, aldehyde products of tissue-damaging lipid peroxidation also increased significantly more in the plasma of spinal cord injury subjects than in trauma controls (2.6 fold vs. 1.9-fold). Spinal cord injury causes a particularly intense systemic inflammatory response. Limiting this response briefly after cord injury should protect the spinal cord and tissues/organs outside the CNS from secondary damage. © 2008 Elsevier Inc. All rights reserved.
Introduction Primary spinal cord injury (SCI) is followed by secondary damage at the injury site, in which inflammation and a large intraspinal influx of leukocytes play an important role (Blight 1992; Fleming et al., 2006; Saville et al., 2004; Taoka and Okajima 1998; Tator and Fehlings 1991). In response to SCI, the release of leukocytes from marginal pool stores and the bone marrow into the circulation increases the number of cells available to cause damage (Furlan et al., 2006). The state of these leukocytes will impact on their actions upon entering the injured cord.
Abbreviations: ANOVA, analysis of variance; COX-2, cyclooxygenase-2; CNS, central nervous system; DHR123, dihydrorhodamine123; DCFH-DA, dichlorofluorescin diacetate; DCF, dichlorofluorescein; fMLP, formyl-methionyl-leucyl-phenylalanine; gp91phox, the catalytic subunit of NADPH oxidase; iNOS, inducible nitric oxide synthase; KPBS, potassium phosphate-buffered saline; MDA, malondialdehyde; MPO, myeloperoxidase; NADPH, nicotinamide adenine dinucleotide phosphate; R123, rhodamine123; RCF, relative centrifugal force; TBA, thiobarbituric acid; SCI, spinal cord injury; TBARS, thiobarbituric acid reactive substances. ⁎ Corresponding author. Fax: +1519663 3789. E-mail address: [email protected] (L.C. Weaver). 0014-4886/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2008.10.022
Leukocytosis is part of the well-known systemic inflammatory response to physical trauma; leukocyte activation consists of increased oxidative activity, phagocytic behaviour and migration (Bhatia et al., 2005; Shih et al., 1999; Tanaka et al., 1991). This inflammatory response can exacerbate the primary cord injury and also cause bystander damage to organs and tissues unaffected by the original injury (Baskaran et al., 2000; Bhatia et al., 2005; Utagawa et al., 2008). Animal studies have shown that CNS injury causes a more intense systemic inflammatory response than general trauma, with ensuing damage to organs such as the lungs and liver (Campbell et al., 2003; Gris et al., 2008). Although inflammation is a recognized source of secondary damage to the injured cord (Fleming et al., 2006), the intensity and character of a systemic inflammatory response to human SCI have not been studied. What are the destructive characteristics of activated circulating inflammatory cells? They may upregulate expression of oxidative enzymes, proteases and other potentially damaging molecules such as pro-inflammatory cytokines (Wang et al., 1997). Indeed, rat studies have confirmed that SCI causes a neutrophilia and changes the character of neutrophils, priming them for greater oxidative burst
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
activity (Gris et al., 2008). An increased capacity for oxidative burst in neutrophils, within hours of SCI in rats, can be induced by circulating pro-inflammatory cytokines and chemokines (Campbell et al., 2005; Wang et al., 1997), as well as by molecules such as platelet activating factor (Botha et al., 1996). Upon entry into the injured cord, these leukocytes would be primed to produce an oxidative burst, releasing proteolytic enzymes and inducing or extending the damage. Although the time-course and other characteristics of the inflammatory leukocyte infiltration into the injured human spinal cord have been studied (Fleming et al., 2006), the impact of SCI on these cells while still in the circulation is less well understood. Using flow cytometry, western blotting, biochemical analyses and immunohistochemical staining, we assessed the presence of reactive oxygen species, expression of a pro-inflammatory transcription factor, expression/activity of oxidative enzymes, and the concentration of products of cell membrane damage (lipid peroxidation) in blood samples obtained from SCI patients. Results were compared to those from a control group who suffered orthopaedic trauma with no CNS injury and from uninjured able-bodied subjects. Materials and methods Patient enrollment These studies were approved by the University of Western Ontario Research Ethics Board for the Review of Health Sciences Research Involving Human Subjects. Venous blood samples were obtained from twenty-five subjects after consent was obtained according to the Declaration of Helsinki. Exclusion criteria were polytrauma, a personal or family history of peripheral neuropathy or autoimmune disease, significant cognitive limitations, malignant cancer within five years of the study, chronic liver disease, regular medication with antiinflammatory drugs, diseases of the blood, pregnancy or lactation. None of the subjects were treated with steroids or other antiinflammatory drugs during the two-week period of the study.
309
The blood samples were taken, when possible, at 6, 12, 24, 48 and 72 h after the injury in trauma controls or spinal cord injury, as well as at 1 week and 2 week after the injuries. If a subject enrolled in the study required a blood transfusion, then no further samples were obtained from that patient. The SCI subjects were diagnosed by magnetic resonance imaging. The trauma controls had significant fractures of vertebrae or long bones without CNS injury. All SCI and trauma control subjects were thoroughly examined to exclude multisystem trauma and had computerized tomography scans of the thorax, abdomen and pelvis to exclude other associated injuries. The patients showed no evidence of systemic infection or local infection during the 2 week period of the study. All blood samples were taken under sterile condition and were analyzed in a laminar flow hood in a tissue culture laboratory. Details of these patients are in Table 1. Oxidative activity analysis Oxidative activity was detected by staining with dihydrorhodamine123 (DHR123), a non-fluorescent agent that is converted by cellular oxidation to the fluorescent dye rodamine123 (R123). Heparinized whole blood samples (200 μl) from uninjured, trauma control and SCI subjects were incubated with DHR123 (1.0 μM) in 800 μl RPMI Medium1640 at 37 °C in sterile conditions; the leukocytes were then isolated by ammonium chloride lysis of red blood cells. At least 50,000 events were analyzed for each blood sample. Neutrophils, monocytes and lymphocytes were initially gated by their characteristic forward and side scatter profiles, which represent size and granularity of the cells, respectively. Cells in these gates were then analyzed for fluorescence intensity. Cell-associated R123 fluorescence within the three populations was determined using a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA) and analyzed by FlowJo software (Tree Star, Ashland, OR). Background fluorescence (in samples incubated without R123) was subtracted from total fluorescence to obtain the value of oxidative activity from each individual sample.
Table 1 Description of subjects Subject Group
Spinal Level(s)
Case #
Age
Sex
Nature of accident
SCI
Trauma control
Uninjured
C5 T12 L1 T12 L1 C5-C7 T12 C6 L1 Fracture location T7-T10 C6-C7 Multiple Multiple Multiple Multiple
P#1 P#2 P#4 P#5 P#6 P#7 P#10 P#13 P#14
41 47 47 33 20 20 44 87 57
M F M M F M M M M
MVA Fall MVA MVA MVA MVA MVA Fall MVA
P#3 P#8 P#9 P#11 P#12 P#15 #1 #2 #3 #4 #5 #6 #7 #8 #9 #10
33 47 25 54 22 21 60 30 35 30 60 55 35 35 35 55
M M M F M F F M M F M M M M M M
MVA Fall MVA MVA MVA MVA
Initial assessmenta
Follow-up assessment
AIS
Motor Score
AIS
Motor score
Months post-SCI
A A A A C C A C A
10 50 52 50 60 40 50 53 60
–b B D A C D B D A
– 53 75 50 77 80 50 74 71
– 10 18 18 15 16 12 6 5
AIS, American Spinal Injury, Association Impairment Scale; C, Cervica; L, Lumbar; MVA, Motor Vehicle Accident; SCI, Spinal Cord Injury; T, Thoracic; a, At admission; b, Patient deceased at 60 h after SCI.
310
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
Myeloperoxidase and DCFH-DA assays Neutrophils express abundant MPO (Bao et al., 2004; Taoka et al., 1997) generating damaging hypochlorous acid. Macrophages also express MPO. Activity of the oxidative enzyme myeloperoxidase (MPO) was analyzed in leukocyte homogenates as described previously (Bao et al., 2004). Reactive oxygen species in the homogenates were detected by their oxidative conversion of the non-fluorescent 2′7′-dichlorofluorescein-diacetate (DCFH-DA) to the fluorescent 2′-7′dichlorofluorescein (DCF) (Bao et al., 2005). Immunocytochemical staining of blood smears A smear was made from blood of each sample and fixed in acetone overnight. The slides containing the smears were kept at −20°C until they were used. Blood smears were immunostained with antibodies to the oxidative enzymes inducible nitric oxide (iNOS, 1:2000, Calbiochem, San Diego, CA) and cyclooxygenase-2 (COX-2,1:500, Rockland, Gilbertsville, PA), to the transcription factor Nuclear Factor-κB p65 (NF-
κB, 1:1000, Abcam Inc, Cambridge MA), and to the catalytic subunit of nicotinamide adeninedinucleotide phosphate (NADPH) oxidase (gp91phox, 1:500, Upstate biotechnology, Lake Placid, KY) using methods described previously (Bao et al., 2004). NF-κB is activated by free radicals, inducing inflammatory gene transcription and cytokine secretion (Tak and Firestein 2001). gp91phox catalyzes the production of superoxide anions (Bao et al., 2004). All staining was performed using the same protocol. After quenching endogenous peroxidase activity, the smears were incubated overnight with the primary antibodies described above. They were then washed in PBS and incubated with biotinylated donkey anti-rabbit or donkey antimouse IgG (1:400) for 2 h. After rinsing in PBS, avidin-biotin complex (1:400) was applied for 2 h. The sections were then washed in PBS and immunoreactivity was visualized after incubation in a glucosediaminobenzidine-nickel solution for 5 min. The stained smears were rinsed in PBS, dehydrated through a gradient of ethanol, cleared, and coverslipped with Cytoseal mountant. PBS was substituted for the primary antibody on control smears in each reaction. A portion of the blood sample from four uninjured subjects was activated ex vivo with
Fig. 1. Oxidative burst in neutrophils and monocytes of uninjured subjects and of trauma controls or spinal cord injury subjects. (A) Example of gating of a blood sample for flow cytometry using physical characteristics of granularity (side scatter) and size (forward scatter). Neutrophils constituted the largest population whereas monocytes made up a smaller population. The lymphocyte population was not analyzed in this study. (B) Histograms plotting the number of cells vs. intensity of R123 fluorescence (log scale) produced by incubation of blood samples with DRH from uninjured subjects (U, orange lines) and from trauma controls (TC, red lines) and spinal cord injury (SCI, brown lines) subjects at 6 h, 24 h and 1 week after injury. Samples incubated without DHR revealed background fluorescence for uninjured (black line), trauma control (light green line) and SCI (blue line) subjects. Trauma control subjects had small increases in fluorescence due to oxidative burst in neutrophils and monocytes. Both neutrophils and monocytes had substantial oxidative activity at 24 h and 1 week after SCI.
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
the pro-inflammatory chemotactic peptide methionyl-leucyl-phenylalanine (fMLP (Gay et al., 1984), 2.5 nM) for 30 min at room temperature (24 °C) prior to preparing the smear to ascertain responses to a known inflammatory agent. For quantitative analysis of optical density of immunoreactive product, three smears from each subject (four subjects each group) were used and a total of ninety neutrophils was analyzed using Image Pro Plus software (Media Cybernetics, Bethesda, MD).
311
(ANOVA) and post-hoc Fisher's protected t-tests (Sokol and Rohlf 1981) or by a Student's t test when only two groups were compared. Sequential data collected during the first 72 h after SCI were compared to those of the uninjured subjects in a separate analysis from the values collected at 1 and 2 weeks after SCI, due to the smaller number of subjects available in the 1- and 2-week periods. Statistical significance was established at P ≤ 0.05. Results
Western blotting to detect iNOS and COX-2 and gp91phox expression Leukocyte counts and oxidative burst after SCI Standard methods for immunoblotting and analysis were used to detect iNOS, COX-2 and gp91phox, employing the antibodies listed above for immunocytochemistry (Bao et al., 2004). The leukocyte homogenates contained mostly neutrophils and lymphocytes, and a smaller number of monocytes, as confirmed by the flow cytometry studies above. As the oxidative activity of lymphocytes is small in comparison to that of neutrophils (Rothe, 1990), the findings in the assays of the leukocyte homogenates can be attributed mostly to neutrophils. Measurement of lipid peroxidation Lipid peroxidation products released from damaged tissues into the plasma were estimated by determining the relative levels of malondialdehyde (MDA), an aldehyde indicator of lipid peroxidation that was measured using a thiobarbiturate reactive substances (TBARS) assay (Bao et al., 2004). Statistical analysis Data are expressed as the mean ± standard error (SE). Differences among groups were established using one-way analysis of variance
The total leukocyte counts at 3.5 ± 1 h after injury in SCI (17.1 ± 1.4 × 109 cells/L) or trauma control (14.7 ± 1.2× 109 cells/L) subjects were greater than normal (4–10 × 109 cells/L). This increase was caused by a neutrophilia as neutrophil counts at this time after SCI were 14.8 ± 1.3 × 109 cells/L and in trauma controls they were 12.6 ± 0.9× 109 cells/L, compared to normal (2–7.5× 109 cells/L). In contrast, the monocyte counts in SCI and trauma controls (0.8± 0.1 × 109 and 0.7 ± 0.1 × 109 cells/L, respectively) were at the upper limit of normal (0.2–0.8× 109 cells/L). To assess oxidative burst, neutrophils, monocytes and lymphocytes were gated, using flow cytometry, by their characteristic forward and side scatter profiles (Fig. 1A) and then cells in each respective gate were analyzed for fluorescence intensity. Individual cases representative of the mean changes in the groups are shown in the histograms of Fig. 1B. After SCI the oxidative activity of neutrophils and monocytes was significantly increased by 4–6 fold in comparison to that of uninjured subjects at 12–48 h and 1 week post-injury (Figs. 2A and B; for details of statistics see Supplemental Table 1) and the maximal increases was at 24 h in neutrophil (6.4 fold). At 12 h, 24 h and 1w after SCI, the oxidative activity of neutrophils and monocytes of SCI subjects was significantly greater than that of trauma controls (Figs. 2A and B).
Fig. 2. Mean and individual changes in oxidative activity in neutrophils (A, C) and monocytes (B, D) after injury in trauma controls and spinal cord injury subjects. A, B. Oxidative activity is expressed as mean ± SE values of R123 fluorescence (after subtraction of background fluorescence) and samples are plotted at times encompassing 6–72 h and 1 and 2 weeks after injury. Uninjured subjects (U, open bars) had low oxidative activity. Oxidative burst in trauma control (TC, grey bars) subjects was slightly but insignificantly increased in circulating neutrophils (A), and monocytes (B). In contrast, oxidative activity increased significantly at 12, 24, 48 h and 1 week after spinal cord injury (SCI, black bars) in neutrophils (A) and monocytes (B). Increases after SCI were greater than those of trauma control subjects in neutrophils and monocytes at 12 h, 24 h and 1 week after injury. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t tests. #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t test. Statistical details in Supplemental Table 1. U: n = 7. At 6, 12, 24, 48, 72 h and 1 and 2 weeks after injury, TC: n = 5, 6, 6, 6, 5, 4 and 4; SCI: n = 4, 9, 9, 8, 5, 6 and 4. C, D. Mean fluorescence intensity (oxidative burst) in neutrophils and monocytes of seven individual uninjured subjects (open circles), three SCI subjects with cervical (C) injury (C5–7, open squares) and six subjects with high thoracic or lumbar injury (T–L, closed squares) are plotted as a scattergram. The responses did not differ according to segmental level of injury.
312
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
The oxidative activity of neutrophils and monocytes from trauma control subjects showed tendencies toward an increase that did not achieve statistical significance. However, when data from all seven sampling times were averaged, oxidative activity of neutrophils and monocytes of trauma control subjects was significantly greater than that of uninjured subjects by 52% and 55%, respectively (P = 0.047 and 0.025, see Supplemental Table 1). Lymphocytes had no evidence of oxidative burst activity when R123 fluorescence from SCI and trauma control subjects was compared to that of uninjured subjects (P = 0.421, data not shown). Of the nine SCI subjects, three had cervical injuries and six had injuries at either the 12th thoracic or 1st lumbar segment. We plotted the oxidative activity of the neutrophils and monocytes of the three subjects with the cervical injury separately from that of the six subjects with low thoracic or lumbar injuries to determine the distribution of response intensity with respect to injury location,. No segment-based pattern of intensity of oxidative burst was apparent in this scattergram (Figs. 2C and D). Particularly at the times (12–48 h) when the sample sizes were greatest, the responses within these subgroups of subjects greatly overlapped. Most responses fell outside the range of oxidative activity of the uninjured control subjects. Free radical formation in leukocytes
Fig. 3. Free radical production and MPO activity in leukocyte homogenates. (A) The presence of free radicals in the leukocytes was estimated by the conversion of DCFH to DCF in homogenates from uninjured (U) subjects and from trauma controls (TC) and spinal cord injury (SCI) subjects at times ranging from 6 h to 2 weeks after injury. DCF concentrations increased significantly in trauma controls and SCI subjects at most times assessed. Changes in leukocytes from SCI subjects were greater than those from trauma control subjects at 12 h, 24 h, 1 week and 2 weeks after injury. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t test. ##P ≤ 0.01; #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t test. U: n = 6. At 6, 12, 24, 48, 72 h and 1 and 2 weeks after injury, TC: n = 5, 6, 6, 6, 5, 4 and 4; SCI: n = 4, 7, 7, 7, 5, 6 and 4. (B) MPO activity increased at 6 h and 12 h after injury in trauma controls. In contrast, MPO activity increased significantly at all times assessed after SCI. The MPO activity at 24 h, 48 h, 72 h and 1 week after SCI was significantly greater than that in trauma control subjects. Definitions of ⁎⁎, ⁎, ##, # and number of subjects/group are as in (A). Statistical details in Supplemental Table 2.
The concentration of fluorescent DCF (the indicator of free radicals) was low in uninjured subjects (Fig. 3A). After injury, the DCF in leukocytes of both trauma controls and SCI subjects increased significantly from 6 h to 2 weeks after injury (P b 0.001, see Supplemental Table 2). Maximal changes were at 24 h after injury when DCF was increased by 1.6 fold in the trauma controls and by 2fold in the SCI subjects. The DCF concentrations were significantly greater in leukocytes of SCI subjects than in trauma controls at 12 h, 24 h and 1 week and 2 weeks after the injury. When data from all seven sampling times were averaged, DCF concentration due to free radical formation was significantly greater in SCI (by 20%) than in trauma control subjects (P = 0.006). Myeloperoxidase activity in leukocytes Samples from uninjured subjects exhibited low-level basal MPO activity (Fig. 3B). In the trauma controls MPO activity increased only at
Fig. 4. Western blot assay of (A) gp91phox, (B) iNOS and (C) COX-2 enzyme expression in leukocyte homogenates from uninjured (U), trauma controls (TC) and spinal cord injury (SCI) subjects at 24 h after injury. (A-C) Upper blots are typical western blots of enzyme expression in each group. Lower blots are β-actin expression demonstrating equal protein loading of gels for each enzyme. Bottom bar graphs show quantification of expression by densitometry, with enzyme expression expressed as a percent of β-actin density ± SE (n = 4 all groups). gp91phox, iNOS and COX-2 expression increased in trauma control and SCI subjects when compared to that of uninjured subjects. gp91phox and iNOS expression was greater after SCI than in trauma control subjects. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t test. ##P ≤ 0.01; #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t test. Statistical details in Supplemental Table 3.
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
6 h and 12 h after injury (P b 0.001, see Supplemental Table 2). In contrast, after SCI, MPO activity increased significantly from 6 h to 2 weeks after SCI. Maximal increases in MPO activity occurred at 6 h
313
after injury in trauma controls (1.7-fold increase) and at 24 h after SCI (2.3-fold increase). The MPO activity in the SCI subjects was significantly greater than that in trauma controls from 24 h to
Fig. 5. Expression of NF-κB, gp91Phox, iNOS and COX-2 in leukocytes of blood smears from uninjured (U), uninjured with fMLP activation (U+fMLP) and trauma controls (TC) and spinal cord injury (SCI) subjects at 24 h after injury. Blood smears were immunostained for the NF-κB (nucleus) and the oxidative enzymes (cytoplasm). Photomicrographs at the top of A-D are typical examples of immunostained cells. Intensely stained cells had morphology characteristic of neutrophils (see text). Calibration bars are 10 μm and apply to all panels. Ninety cells in total were analyzed by ImagePro software to obtain a measure of intensity of staining per cell (relative optical density/area). Bar graphs in lower panels of A–D depict mean relative optical density/area ± SE for each group (n = 4 subjects/group). Black and white arrow heads indicate the lobulated nucleus. After treatment of blood samples with fMLP and in trauma control and SCI subjects, expression of NF-κB, gp91Phox, iNOS and COX-2 all increased significantly when compared to values from uninjured subjects. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t test. ##P ≤ 0.01, significantly different from trauma controls by Fisher's protected t test, +P = 0.07, tended to differ from trauma controls by this test. Statistical details in Supplemental Table 4.
314
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
1 week after injury. Averaged activity from the seven sampling times also showed that MPO activity in SCI subjects was significantly greater (by 74%) than in trauma controls (P b 0.001). Expression of oxidative enzymes and NF-κB in leukocytes gp91phox protein expression was readily detected in leukocytes of uninjured subjects (Fig. 4A). At 24 h after injury in trauma control or SCI subjects, this gp91phox expression increased significantly (1.5-fold and 1.8-fold, respectively) compared to the uninjured subjects and expression was significantly greater in SCI subjects than in trauma controls (P b 0.001, see Supplemental Table 3). Expression of iNOS increased from the low levels in the uninjured subjects to significantly greater expression at 24 h after either type of injury (Fig. 4B, P b 0.001). Again, increases after SCI (2.2-fold) were greater than those in trauma controls (1.8-fold). Finally, expression of COX-2 also increased significantly in leukocytes of trauma control (4.2-fold) or SCI (4.9fold) subjects when compared to the lower levels in uninjured subjects (Fig. 4C, P b 0.001). The robust increases in COX-2 expression after either type of injury were not significantly different from each other. Immunocytochemical staining of leukocytes in blood smears confirmed expression of these oxidative enzymes and of the proinflammatory transcription factor NF-κB at 24 h after injury and revealed the phenotype of cells expressing these molecules. Expression was consistently found in neutrophils identified by their size (diameter 10–12 μm) and lobular nucleus. Cells with morphology of monocytes or lymphocytes exhibited low levels of expression of each of these proteins (data not shown). Little expression of NF-κB was detected in neutrophils from uninjured subjects (Fig. 5A top panels). In contrast, expression of NF-κB was clear after incubation of samples from uninjured subjects after fMLP stimulation and at 24 h after injury in trauma controls and SCI subjects. Active NF-κB was mostly in the nucleus with fainter expression in the cytoplasm (Fig. 5A top panels). Quantification of expression by densitometry revealed significant, approximately 2-fold increases after treatment with fMLP, as well as in samples from trauma controls and SCI subjects, when compared to that in the uninjured subjects (Fig. 5A lower panel, P b 0.001, see Supplemental Table 4). Expression of gp91phox, iNOS and COX-2 in neutrophils was confined to the cytoplasm and appeared at low levels even in leukocytes of uninjured subjects (Figs. 5B–D top panels). Treatment of blood samples from uninjured subjects with fMLP significantly increased expression of gp91phox, iNOS and COX-2 by 1.6-fold, 1.6-fold and 2fold, respectively (P b 0.001). In trauma controls, expression of these three enzymes increased significantly by 1.5-fold, 1.8-fold and 2-fold, respectively, when compared to that in uninjured subjects (Figs. 5B–D, lower panels). Likewise, after SCI, expression of gp91phox, iNOS and COX-2 increased by 1.8-fold, 2.7-fold and 2.3-fold, respectively. Increases in iNOS expression in leukocytes of SCI subjects were significantly greater than those in trauma controls. The tendency for increased gp91phox expression in SCI subjects compared to trauma controls almost attained statistical significance (P = 0.07). Lipid peroxidation The TBARS concentration (revealing lipid peroxidation products) was low in the plasma of uninjured subjects (Fig. 6). In trauma controls, the plasma TBARS concentration increased significantly above that in uninjured subjects at 24–72 h and at 2 weeks after the injury (P = 0.022, 6–72 h; P = 0.018, 1, 2 weeks, see Supplemental Table 5). The maximal (1.9-fold) increase occurred at 24 h after injury. In contrast, the plasma TBARS concentration increased significantly from 6 h to 2 weeks after SCI, when compared to values in uninjured subjects. Again maximal increases (2.6-fold) occurred at 24 h after injury. At 12–48 h and 1 week after injury, the plasma TBARS concentration in SCI subjects was
Fig. 6. Mean changes in products of lipid peroxidation detected in plasma samples obtained from uninjured (U), trauma controls (TC) and spinal cord injury (SCI) subjects. The products were primarily malondialdehye, detected by the thiobarbituric acid reactive substances (TBARS) assay, and expressed as TBARS concentration/mg protein (mean ± SE) at times from 6–72 h and 1 and 2 weeks after injury. Lower levels of TBARS were present in the serum of uninjured subjects and this concentration increased significantly at 24–72 h and 2 weeks in trauma controls. In contrast, TBARS concentrations increased significantly at all times measured after SCI and were significantly higher than those in trauma controls at 12–48 h and 1 week after injury. ⁎⁎P ≤ 0.01; ⁎P ≤ 0.05, significantly different from uninjured by Fisher's protected t tests. ##P ≤ 0.01; #P ≤ 0.05, significantly different from trauma controls by Fisher's protected t tests. Statistical details in Supplemental Table 5. U: n = 6. At 6, 12, 24, 48, 72 h and 1 and 2 weeks after injury, TC: n = 5, 6, 6, 6, 5, 4 and 4; SCI: n = 4, 7, 7, 7, 5, 6 and 4.
significantly greater than in trauma controls (Fig. 6). When data from all seven sampling times were averaged, the plasma concentration of TBARS in SCI subjects was significantly greater than that in trauma controls (P b 0.0001). This measure of lipid peroxidation in the subjects with cervical injury (C5–7, n = 3) at 24 h and 48 h after injury (335 ± 11 and 293 ± 11 nmol/100 mg protein) was not different from that in subjects with low thoracic or lumbar injury (T12-L1, n = 4) SCI (359 ± 75 and 385 ± 107 nmol/100 mg protein). Discussion Following traumatic SCI, neutrophils and then monocyte/macrophages infiltrate the lesion from the circulation (Fleming et al., 2006). Our study reveals that insult to the spinal cord caused a particularly intense induction of oxidative burst and associated enzymes in the circulating leukocytes, significantly greater than that in the trauma control subjects. The highly activated neutrophils and monocytes after SCI could play a significant role in the secondary damage that occurs within and outside the injured spinal cord. The different parameters measured in our study correlated well with each other. Increased oxidative burst synchronized with increased free radical production, MPO activity and expression of the key enzymes in the induction of oxidative burst (gp91phox, iNOS and COX-2). These changes paralleled evidence of secondary tissue damage due to lipid peroxidation. Expression of the pro-inflammatory transcription factor NF-κB also increased. As neutrophils are the first leukocytes to enter the injured human spinal cord (Fleming et al., 2006), their robust oxidative activity likely contributes to the early stages of secondary damage to surviving tissue. We used several different methods to detect changes in leukocyte oxidative activity after SCI. First, using flow cytometry we found that the neutrophils and monocytes increased their oxidative activity intensely in the first two days after the injury. Lymphocytes did not undergo oxidative burst activity. At the peak time of oxidative activity (24 h after SCI), we extended the investigation to examine oxidative enzymes and free radical production using western blot and biochemical analysis from leukocyte homogenates. As monocytes constitute only a small portion of the white cell population, our data from the leukocyte homogenates likely reflect the responses of neutrophils. Similarly, few monocytes were detected in the blood
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
smears. The increased oxidative activity may well have been present in the monocytes but, with the exception of the flow cytometry experiments, we did not isolate this small population and they likely made a small contribution to the responses that we measured. Studies in rats and mice (Blight 1992; Saville et al., 2004) have shown that neutrophils generate reactive oxygen and nitrosyl radicals as well as a variety of damaging oxidative enzymes (Bao et al., 2004; Bao et al., 2005). Neutrophils also produce pro-inflammatory cytokines and chemokines (Brandt et al., 2000; Cassatella et al., 1995). Cytokine and chemokine gene expression in neutrophils depends on activation of the pro-inflammatory transcription factor NF-kB (McDonald et al., 1997) and the NF-κB activation in circulating neutrophils in our study suggests that these cells were producing proinflammatory and oxidative proteins. Ours is the first demonstration of activated NF-kB in circulating neutrophils after traumatic SCI. Induction of hepatic NF-kB recently has been shown to play a key role in activation and recruitment of neutrophils into the injured mouse brain (Campbell et al., 2008). A robust systemic inflammatory response to severe trauma or infection often culminates in multiple organ failure (Moore et al., 1994). The multiple organ dysfunction after severe general trauma in humans, such as that caused by multiple bone fractures or extensive burns, is mediated, in part, by neutrophils that become activated and more responsive to stimulation (Bhatia et al., 2005). For example, neutrophils infiltrate the lungs in conditions that produce a systemic inflammatory response such as burns, multiple trauma, and infections in humans and/or animal models (Bhatia et al., 2005; Gris et al., 2008; Sunil et al., 2002). A recent study in rats revealed that a systemic inflammatory response to SCI can cause a major influx of neutrophils into the lungs, leading to oxidative and proteolytic enzyme expression/activity within the organ (Gris et al., 2008). This results in lung damage, revealed by the presence of lipid peroxidation. Brain and spinal cord injury also causes an influx of leukocytes into the liver and causes liver damage, due to the almost immediate triggering of hepatic chemokine and acute phase protein expression and release (Campbell et al., 2005; Wilcockson et al., 2002). Such systemic inflammation likely contributes to organ dysfunction after SCI, as it does after brain injury and other traumatic insults (Baskaran et al., 2000; Bhatia et al., 2005; Ott et al., 1994). Therefore, the activation of leukocytes by SCI, and downstream consequences of this activation, not only exacerbates the spinal injury, but also puts key organs such as the lungs and liver at risk. Indeed respiratory failure is the leading cause of short-term and long-term mortality after human SCI (DeVivo et al., 1999) and, in our small sample of subjects, one died of respiratory failure and a second experienced severe pulmonary complications. The most important finding in our study was that the systemic inflammatory response after SCI was usually greater than the response to severe bone fractures. The difference was not in the degree of soft tissue or bone damage but instead appears to correlate with injury to the CNS. The mechanism for this intense reaction is unknown but brain injury also causes a particularly intense systemic inflammatory response (Campbell et al., 2005). After direct CNS injury, loss of crucial feedback control of immune function occurs. For example, significant cord injury at all levels releases opiates (Bakshi et al., 1990) that can cause neutrophil activation (Menzebach et al., 2003), contributing to the systemic inflammatory response. In addition, injury to the upper thoracic or cervical cord severely depresses sympathetic activity and catecholamine levels (Maiorov et al., 1997; Mathias et al., 1975). As catecholamines inhibit neutrophil oxidative burst and suppress expression of integrins (Trabold et al., 2007), the low catcholamine levels after SCI also are proinflammatory. These effects of CNS injury on immune function, would not occur after general trauma. The systemic inflammatory response that we report in the SCI and trauma control subjects is a well-known response to trauma. However, a significant literature also reports immunosuppression after trauma,
315
including SCI (Campagnolo et al., 1997; Cruse et al., 2000; Meisel et al., 2005). This immunosuppression primarily affects the function of natural killer cells, T-cells, B cells and macrophages in humans and animals at days to weeks after injury (Cruse et al., 1992; Cruse et al., 2000; Furlan et al., 2006; Lucin et al., 2007; Riegger et al., 2007). Unlike the responses in our study, the depressant effects of SCI on immune function are mostly time-dependent with greater suppressive effects in the chronic phases of injury. Immunosuppression also can very with the level of SCI, with cervical or high thoracic injuries having greater effects on lymphocytes and phagocytic properties of neutrophils (Campagnolo et al., 1997; Lucin et al., 2007; Meisel et al., 2005). This response is independent of time after SCI. In contrast, the intensity of oxidative burst and lipid peroxidation in our study did not vary with the segment of injury. In conclusion, SCI causes a profound systemic inflammatory response that may have widespread consequences. This response within the circulation is a ready target for intravenous intervention with agents that transiently limit the migration of neutrophils and monocytes into the injured spinal cord or visceral organs. Acknowledgments We especially thank all of the volunteers and patients for contributing to this study. We appreciate that the SCI and trauma subjects made the decision to contribute to this research, and permitted the blood samples to be drawn, under circumstances when their own personal needs were very great. We also thank all of the fellows, nurses and staff at the University Health Sciences Centre Victoria Hospital Campus and Robarts Research Institute who facilitated our work by coordinating the timely acquisition and transport of samples to our laboratory. We appreciate the advice of our flow cytometry manager, Dr. Kristin Chadwick, assistance with the assays by summer student Ms. Maria Pineda, and data preparation by Ms. Eilis Hamilton. The authors appreciate the constructive critique of the manuscript by Dr. Canio Polosa. This research was funded by a New Emerging Team grant from the Canadian Institutes of Health Research (RMN63187). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.expneurol.2008.10.022. References Bakshi, R., Newman, A.H., Faden, A.I., 1990. Dynorphin A-(1–17) induces alterations in free fatty acids, excitatory amino acids, and motor function through an opiatereceptor-mediated mechanism. J. Neurosci. 10 (12), 3793–3800. Bao, F., Chen, Y., Dekaban, G.A., Weaver, L.C., 2004. Early anti-inflammatory treatment reduces lipid peroxidation and protein nitration after spinal cord injury in rats. J. Neurochem. 88, 1335–1344. Bao, F., Dekaban, G.A., Weaver, L.C., 2005. Anti-CD11d antibody treatment reduces free radical formation and cell death in the injured spinal cord of rats. J. Neurochem. 94, 1361–1373. Baskaran, H., Yarmush, M.L., Berthiaume, F., 2000. Dynamics of tissue neutrophil sequestration after cutaneous burns in rats. J. Surg. Res. 93, 88–96. Bhatia, R.K., Pallister, I., Dent, C., Jones, S.A., Topley, N., 2005. Enhanced neutrophil migratory activity following major blunt trauma. Injury 36, 956–962. Blight, A.R., 1992. Macrophages and inflammatory damage in spinal cord injury. J. Neurotrauma 9 (Supp 1), S83–S91. Botha, A.J., Moore, F.A., Moore, E.E., Peterson, V.M., Silliman, C.C., Goode, A.W., 1996. Sequential systemic platelet-activating factor and interleukin 8 primes neutrophils in patients with trauma at risk of mulitple organ failure. Br. J. Surg. 83, 1407–1412. Brandt, E., Woerly, G., Younes, A.B., Loiseau, S., Capron, M., 2000. IL-4 production by human polymorphonuclear neutrophils. J. Leukocyte Biol. 68, 125–130. Campagnolo, D.I., Bartlett, J.A., Keller, S.E., Sanchez, W., Oza, R., 1997. Impaired phagocytosis of Staphylococcus aureus in complete tetraplegics. Am. J. Phys. Med. Rehabil. 76 (4), 276–280. Campbell, S.J., Hughes, P.M., Iredale, J.P., Wilcockson, D.C., Waters, S., Docagne, F., Perry, V.H., Anthony, D.C., 2003. CINC-1 is an acute-phase protein induced by focal brain injury causing leukocyte mobilization and liver injury. FASEB J. 17, 1168–1170.
316
F. Bao et al. / Experimental Neurology 215 (2009) 308–316
Campbell, S.J., Perry, V.H., Pitossi, F.J., Butchart, A.G., Chertoff, M., Waters, S., Dempster, R., Anthony, D.C., 2005. Central nervous system injury triggers hepatic CC and CXC chemokine expression that is associated with leukocyte mobilization and recruitment to both the central nervous system and the liver. Am. J. Pathol. 166, 1487–1497. Campbell, S.J., Anthony, D.C., Oakley, F., Carlsen, H., Elsharkawy, A.M., Blomhoff, R., Mann, D.A., 2008. Hepatic nuclear factor kappa B regulates neutrophil recruitment to the injured brain. J. Neuropathol. Exp. Neurol. 67 (3), 223–230. Cassatella, M.A., Gasperini, S., D'Andrea, A., Ma, X., Trinchieri, G., Meda, L., 1995. Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25, 1–5. Cruse, J.M., Lewis, R.E., Bishop, G.R., Kliesch, W.F., Gaitan, E., 1992. Neuroendocrineimmune interactions associated with loss and restoration of immune system function in spinal cord injury and stroke patients. Immunol. Res. 11 (2), 104–116. Cruse, J.M., Lewis, R.E., Dilioglou, S., Roe, D.L., Wallace, W.F., Chen, R.S., 2000. Review of immune function, healing of pressure ulcers, and nutritional status in patients with spinal cord injury. J. Spinal Cord Med. 23 (2), 129–135. DeVivo, M.J., Krause, J.S., Lammertse, D.P., 1999. Recent trends in mortality and causes of death among persons with spinal cord injury. Arch. Phys. Med. Rehabil. 80 (11), 1411–1419. Fleming, J.C., Norenberg, M.D., Ramsay, D.A., Dekaban, G.A., Marcillo, A.E., Saenz, A.D., Pasquales-Style, M., Dietrich, W.D., Weaver, L.C., 2006. The cellular inflammatory response in human spinal cords after injury. Brain 129 (Pt 12), 3249–3269. Furlan, J.C., Krassioukov, A.V., Fehlings, M.G., 2006. Hematologic abnormalities within the first week after acute isolated traumatic cervical spinal cord injury: a casecontrol cohort study. Spine 31, 2674–2683. Gay, J.C., Beckman, J.K., Brash, A.R., Oates, J.A., Lukens, J.N., 1984. Enhancement of chemotactic factor-stimulated neutrophil oxidative metabolism by leukotriene B4. Blood 64, 780–785. Gris, D., Hamilton, E.F., Weaver, L.C., 2008. The systemic inflammatory response after spinal cord injury damages lungs and kidneys. Exp. Neurol. 211, 259–270. Lucin, K.M., Sanders, V.M., Jones, T.B., Malarkey, W.B., Popovich, P.G., 2007. Impaired antibody synthesis after spinal cord injury is level dependent and is due to sympathetic nervous system dysregulation. Exp. Neurol. 207 (1), 75–84. Maiorov, D.N., Weaver, L.C., Krassioukov, A.V., 1997. Relationship between sympathetic activity and arterial pressure in conscious spinal rats. Am. J. Physiol. 272, H625–H631. Mathias, C.J., Christensen, N.J., Corbett, J.L., Frankel, H.L., Goodwin, T.J., Peart, W.S., 1975. Plasma catecholamines, plasma renin activity and plasma aldosterone in tetraplegic man, horizontal and tilted. Clin. Sci. Mol. Med. 49, 291–299. McDonald, P.P., Bald, A., Cassatella, M.A., 1997. Activation of the NF-kappaB pathway by inflammatory stimuli in human neutrophils. Blood 89, 3421–3433. Meisel, C., Schwab, J.M., Prass, K., Meisel, A., Dirnagl, U., 2005. Central nervous system injury-induced immune deficiency syndrome. Nat. Rev. Neurosci. 6 (10), 775–786.
Menzebach, A., Hirsch, J., Hempelmann, G., Welters, I.D., 2003. Effects of endogenous and synthetic opioid peptides on neutrophil function in vitro. Br. J. Anaesth. 91 (4), 546–550. Moore, E.E., Moore, F.A., Franciose, R.J., Kim, F.J., Biffl, W.L., Banerjee, A., 1994. The postischemic gut serves as a priming bed for circulating neutrophils that provoke multiple organ failure. J. Trauma 37 (6), 881–887. Ott, L., McClain, C.J., Gillespie, M., Young, B., 1994. Cytokines and metabolic dysfunction after severe head injury. J. Neurotrauma 11, 447–472. Riegger, T., Conrad, S., Liu, K., Schluesener, H.J., Adibzahdeh, M., Schwab, J.M., 2007. Spinal cord injury-induced immune depression syndrome (SCI-IDS). Eur. J. Neurosci. 25 (6), 1743–1747. Rothe, G.V.G., 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidin and 2´,7´-dichlorofluorescin. J. Leukocyte Biol. 47, 440–448. Saville, L.R., Pospisil, C.H., Mawhinney, L.A., Bao, F., Simedria, F.C., Peters, A.A., O'Connell, P.J., Weaver, L.C., Dekaban, G.A., 2004. A monoclonal antibody to CD11d reduces the inflammatory infiltrate into the injured spinal cord: A potential neuroprotective treatment. J. Neuroimmunol. 156, 42–57. Shih, H.C., Su, C.H., Lee, C.H., 1999. Superoxide production of neutrophils after severe injury: impact of subsequent surgery and sepsis. Am. J. Emerg. Med. 17 (1), 15–18. Sokol, R.R., Rohlf, F.J., 1981. Biometry: The Principles and Practice of Statistics in Biological Research. W.H. Freeman, San Francisco. Sunil, V.R., Connor, A.J., Zhou, P., Gordon, M.K., Laskin, J.D., Laskin, D.L., 2002. Activation of adherent vascular neutrophils in the lung during acute endotoxemia. Respir. Res. 3 (1), 21. Tak, P.P., Firestein, G.S., 2001. NFkB: a key role in inflammatory diseases. J. Clin. Invest. 107, 7–11. Tanaka, H., Ogura, H., Yokota, J., Sugimoto, H., Yoshioka, T., Sugimoto, T., 1991. Acceleration of superoxide production from leukocytes in trauma patients. Ann. Surg. 214 (2), 187–192. Taoka, Y., Okajima, K., 1998. Spinal cord injury in the rat. Prog. Neurobiol. 56, 341–358. Taoka, Y., Okajima, K., Uchiba, M., Murakami, K., Kushimoto, S., Johno, M., Naruo, M., Okabe, H., Takatsuki, K., 1997. Role of neutrophils in spinal cord injury in the rat. Neuroscience 79 (4), 1177–1182. Tator, C.H., Fehlings, M.G., 1991. Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms. J. Neurosurg. 75, 15–26. Trabold, B., Gruber, M., Frohlich, D., 2007. Functional and phenotypic changes in polymorphonuclear neutrophils induced by catecholamines. Scand. Cardiovasc. J. 41 (1), 59–64. Utagawa, A., Truettner, J.S., Dietrich, W.D., Bramlett, H.M., 2008. Systemic inflammation exacerbates behavioral and histopathological consequences of isolated traumatic brain injury in rats. Exp. Neurol. 211 (1), 283–291. Wang, C.X., Olschowka, J., Wrathall, J.R., 1997. Increase of interleukin-1beta mRNA and protein in the spinal cord following experimental traumatic injury in the rat. Brain Res. 759, 190–196. Wilcockson, D.C., Campbell, S.J., Anthony, D.C., Perry, V.H., 2002. The systemic and local acute phase response following acute brain injury. J. Cereb. Blood Flow Metab. 22, 318–326.