Background
Brain is one of the remote organs subjected to injurious effects of severe burns [
1]. Survivors suffering from extensive burn injury present long-term cognitive impairment, including depression, anxiety, post-traumatic stress disorder [
2,
3], and alteration in painful sensation as well as sensory sensitivity in later life [
4]. In animal studies, magnetic resonance imaging has identified marked changes in the brain up to 3 days postburn (pb), most notably swelling and lesions [
5], changes in cerebral blood flow [
6], dysregulation of glucose metabolism [
7], and disruption of the blood-brain barrier (BBB) [
8,
9].
Neuroinflammation is a frequent consequence of sepsis and septic shock [
10]. Approximately 93% of burn patients show clinical signs of a systemic inflammatory response syndrome before succumbing to their injuries [
11], and this syndrome can deteriorate and develop into severe sepsis [
12]. After burn injury, there is a dramatic increase in proinflammatory cytokines in brain as early as 3 hours (h) [
13,
14] and a compromised BBB leading to a large infiltration of macrophages [
9]. Beneficial as well as deleterious effects have been ascribed to immune cells that infiltrate the nervous system after neural injury [
15‐
19]. Despite the correlation between cerebral complications in severe burn victims and mortality, burn-induced neuroinflammation continues to be an underestimated entity in critically ill burn patients [
10].
Gelsolin was first described as a ~90 kDa cytoplasm actin-binding protein with capping and severing activities [
20]. Further studies have confirmed a secreted gelsolin isoform in blood plasma [
21]. Recent reports have documented that it also participates in the regulation of the systemic immune response. Extracellular gelsolin is involved in host immune recognition of bacterial wall molecules during cell division or attack by immune components, while cytoplasmic gelsolin is necessary for macrophage motility in culture, and its absence is likely to impair recruitment of macrophages to a site of crush injury of sciatic nerve [
22]. In fact, overexpression of gelsolin could alter actin dynamics in Jurkat T cells, correlating with inhibition of activation-dependent signaling pathways [
23]. Moreover, cytoplasmic gelsolin depletion is observed in diverse states of inflammation that are associated with tissue injury and actin release, including hemorrhagic shock [
24], early sepsis, trauma, and rheumatoid arthritis [
25]. In addition, its deficiency has been found to correlate with septic mortality [
26] and prognosis [
27], suggesting that gelsolin might play a crucial protective role in the course of sepsis.
Accordingly, gelsolin replacement might be considered as a potential therapy for the lethal condition of sepsis [
28]. It could solubilize circulating actin aggregates and shift expressed cytokines toward an anti-inflammatory profile [
28], resulting in a significant reduction of mortality in endotoxemic mice. Since gelsolin has been shown to significantly blunt neutrophil recruitment to lungs [
29] and to markedly attenuate vascular permeability in burn injury in rats [
30], we hypothesized that, in severe burn injury of mice, a single dose of gelsolin might attenuate neuroinflammation, which might ultimately protect the brain from injurious effects following the acute insult.
Methods
Animal model of burn injury
Male Balb/c mice (20-25 g, 8-9 weeks old, obtained from the Laboratory Animal Institute, Beijing, China) were anesthetized, and the dorsal and lateral surfaces of the mice were shaved. Mice were secured in a protective template on their backs with an opening corresponding to 15% of the total body surface area (TBSA), and the exposed skin was immersed in 95°C water for 8 seconds (s). This procedure has been shown to produce a 15% TBSA full-thickness scald injury. Sham-injured mice were subjected to all of the procedures except that the temperature of the bath was the same as room temperature. Immediately following injury, the mice were dried and allowed to recover under a heating lamp. Both sham- and burn-injured mice received 1.0 ml of fluid for resuscitation intraperitoneally (i.p.) (Ringer's solution). Animals were then housed in individual cages in a temperature and humidity controlled room with 12 hours (h) light and 12 h darkness before being sacrificed. All experimental manipulations were undertaken in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, with the approval of the Scientific Investigation Board of the Chinese PLA General Hospital, Beijing, China.
Intravenous gelsolin infusion
Animals were randomly divided into five groups: intact controls, sham-burn mice, placebo controls that underwent burn injury with an equivalent amount of bovine serum albumin (BSA; Fisher Scientific, Fair Lawn, NJ), and burned mice treated with either a low dose (2 mg/kg, Gsn-L) or a high dose (20 mg/kg, Gsn-H) of recombinant human gelsolin (Sigma-Aldrich, Shanghai, China), according to a previous report [
31], in 0.1 ml of sterile saline
via tail vein immediately after burn injury. Then the animals (9-10 mice per group) were sacrificed at 8, 24, 48 and 72 h postburn (pb). Tissue and plasma samples were collected and stored at -80°C.
Survival rate
Survival rates were recorded for the low- or high-dose gelsolin-treated mice (n = 30 per group), the placebo-treated mice (n = 30), and the sham-injured mice (n = 10) without further intervention. Differences in survival rates among the groups were analyzed by the Kaplan-Meier method using an SPSS software package.
Functions of T lymphocytes
Splenic mononuclear cells (MNC) were separated by Ficoll-Paque density centrifugation and were cultivated in complete RPMI-1640 medium in flat-bottomed 96-well microtitre plates (4 × 105 cells per well) stimulated by the T-cell mitogen concanavalin A (ConA, 5 mg/L; Sigma) for 48 h. Cell-free supernatant fractions were collected and stored at -80°C until analysis for IL-2 by ELISA (ExCell Biology Inc., Shanghai, China). T cell proliferation was examined using a 3-(4, 5-dimethylthiazol -2-yl)- 2, 5-diphenyltetrazolium bromide (MTT) method with absorbance at 450 nm in a multiplate spectrophotometer (Spectra MR; Dynex, Richfield, MN, USA).
Tissue preparation for immunostaining
Mice (3-4 per group) were killed by cervical dislocation and the brains were removed and post-fixed for 24 h in 4% paraformaldehyde solution, followed by 30% sucrose in phosphate buffer saline (PBS) for another 24-48 h. Brains were stored at -80°C until used to prepare frozen sections at 30 μm thickness. These were serially collected in PBS and finally stored in cryoprotectant solution at -30°C. Some of the brain sections were mounted on lysine-coated slides and stained with hematoxylin and eosin (H&E).
Quantitative polymerase chain reaction (PCR)
Brains from the remaining mice (5-6 mice per group) were carefully dissected and collected, snap frozen in liquid nitrogen, and stored at -80°C. Different regions (cortex, hippocampus and striatum) were used for total RNA extraction using a NucleoSpin® RNA II Kit (Macherey-Nagel Inc., PA, USA) following the manufacturer's instructions, and used for cDNA synthesis with Superscript II (Promega, Beijing, China). Real-time PCR amplification was achieved in 25 μl reaction mixtures containing 5 μl of cDNA sample, 12.5 μl of SYBR Green PCR Master Mix (SYBR green; Applied Biosystems, Foster City, CA, USA) and specific primers (SBS Genetech Co. Ltd, Beijing, China). An ABI Prism 7700 sequence detection system (Applied Biosystems) with SYBR-green fluorescence was used for assay. Cycling conditions were a 10-min hot start at 95°C followed by 5 cycles of denaturation steps at 95°C for 40 s, an annealing step at 60°C for 30 s, and an extension temperature at 72°C for 30 s. Each sample was run in triplicate. β-actin was used as housekeeping mRNA to normalize gelsolin transcript abundance. Data were analyzed by using sequence Detector Systems version 2.0 software.
Each sample was tested in triplicate. The relative concentration of mRNA was calculated using the formula x = 2
-ΔΔCt, where x fold change in the target gene at each detection time, normalized to β-actin and relative to the expression of intact mice [
32].
Immunohistochemistry
Sections used for immunocytochemistry were incubated in 0.3% hydrogen peroxide (H2O2) for 10 min, and incubated free-floating in antibodies (Abs) of polyclonal anti-mouse ionized calcium-binding adapter molecule 1 (Iba-1, 1:1000; Wako, Osaka, Japan), monoclonal anti-mouse CD11b (Mac-1, 1:1000; EuroBioScience, Lund, Sweden), monoclonal anti-mouse CD45 (1:1000; EuroBioScience), or rabbit anti-cleaved caspase-3 (1:50; Cell Signaling, Danvers, MA, USA) with 3% normal goat serum, 0.05%Triton-X in PBS, for 24-48 h rotating at 4°C. The tissue was then rinsed in PBS and incubated for 1 h in biotinylated anti-rabbit IgG (1:200; Vector Laboratories, Burlingame, CA, USA), rotating at room temperature. The tissue was then rinsed in PBS and incubated for 1 h in ABC solution (Vector Laboratories). Following incubation, sections were rinsed with PBS for 20 min and were developed by incubating in 0.025% diamino-benzidine (DAB; Sigma-Aldrich) and 0.002% H2O2 in PBS. The DAB reaction was halted using PBS, followed by three 10-min PBS rinses.
Quantification of immunohistochemistry
For quantitative image analysis of periventricular immunostaining, serial sagittal sections of one hemisphere were collected (lateral position +0.5 to +2.25 from Bregma). Iba-1-, CD11b- and CD45-immunostained preparations of sagittal brain sections were evaluated for 4-5 animals from each group. For each animal, antigens were detected in 10 parallel sections having a distance of 70 mm from each other and showing both striatum and cortex. All images were acquired on a BX-61 microscope (Olympus Optical Co., Tokyo, Japan), equipped with a digital camera (F-View II; Olympus Optical Co.). Quantification of immunoreactive cells within the cortex and the striatum was performed at 40 × magnification by a researcher blinded to the treatment. For each animal, average values from all sections were determined.
Neuron-specific enolase (NSE) and soluble protein-100 (S100) detection
Brain tissues were weighed and homogenized after addition of 3 ml/g (1:4) saline with protease inhibitor cocktail (Applygen Technologies Inc., Beijing, China). The supernatants were collected for NSE and S100 analysis in duplicate using available quantitative 'sandwich' enzyme-linked immunosorbent assay kits (Rapidbio, CA, USA). Sensitivity of the assays was 1.0 pg/ml for S100 and 0.1 ng/ml for NSE.
Western blot
The dissected brain tissues were collected, snap-frozen in liquid nitrogen and stored at -80°C. Tissue was homogenized in RIPA buffer with protease inhibitor (Applygen Technologies Inc.). The total amount of protein was determined by bicinchoninic acid protein assay (Applygen Technologies Inc.). Samples (100 μg protein) were separated by 8% SDS-PAGE and electroblotted to nitrocellulose membrane, which were blocked by incubation in 3% (w/v) bovine serum albumin dissolved in TBS-T (150 mM NaCl, 50 mM Tris, 0.05% Tween 20). Following transfer, proteins were probed using a rabbit monoclonal phospho-p44/42 extracellular regulated kinase 1/2 (ERK1/2) (1:2000; Cell Signaling) in TBS-T. Horseradish peroxidase-conjugated secondary Ab was used at a 1:1000 dilution in TBS-T. After extensive washing, protein bands detected by Abs were visualized by ECL reagent (Applygen Technologies Inc.) after exposure on autoradiograph film (Fuji Film; Kodak Scientific Imaging Film, Beijing). Membranes were then stripped and re-probed with p44/42 MAPK (ERK1/2) mouse monoclonal Ab (1:1000; Cell Signaling) to confirm equal protein loading. The films were subsequently scanned, and band intensities were quantified using Image software.
Assessment of cysteinyl aspartate-specific protease (caspase)-3 activity
Caspase-3 activity was measured using a colorimetric assay according to the manufacturer's instructions (BioVision, Mountain View, CA, USA). The brain tissues were lysed in buffer (50 mM HEPES, pH 7.4, 0.1% CHAPS, 1 mM DTT, 0.1 mM EDTA and 0.1% Triton X-100) and centrifuged at 12, 000 × g for 10 min at 4°C. After determination of protein concentration by bicinchoninic acid method (Applygen Technologies Inc.), the cell extract (200 μg of protein) was added to the assay buffer (100 mM HEPES, pH 7.4, 0.1% CHAPS, 10 mM DTT, 10% glycerol, and 2% (v/v) dimethylsulfoxide) containing chromogenic substrates (2 mM) and incubated for 4 h at 37°C. Caspase-3 activity was determined by measuring the absorbance at 405 nm using a microplate reader (Spectra MR; Dynex, Richfield, MN, USA).
Determination of plasma gelsolin concentrations
At 8, 24, 48 and 72 h after burns or sham injury, the animals were anesthetized, and blood obtained by cardiac puncture was placed in a heparinized tube (n = 6 samples each group per time point). The blood was centrifuged and plasma gelsolin concentrations were determined in duplicate with a mouse gelsolin ELISA detection kit (USCN Life, Wuhan, China).
Statistic analysis
All data are expressed as mean ± SD from three or more independent experiments. Statistical comparisons among different groups were done by one-way analysis of variance (ANOVA) with Dunnett's multiple comparison tests using SPSS software (IBM, Beijing, China). Differences with p < 0.05 were considered statistically significant.
Discussion
A successful therapeutic strategy for brain injury should include inhibition of proinflammatory cytokines, promotion of anti-inflammatory cytokines, suppression of autoimmunity to CNS antigens and reduction in recruitment of inflammatory cells, etc. In the present study, we report protective effects of gelsolin in brain of mice subjected to burn injury, characterized by amelioration of pathological lesions and suppression of microglial activation, which might be associated with enhanced recruitment of CD11b+ as well as CD45+ cells. Likewise, gelsolin could substantially down-regulate the marked expression of both early (IL-1β, IL-6) and late proinflammatory cytokines (HMGB1) in the brain. In addition, treatment with gelsolin significantly reduced caspase-3 activity and inhibited ERK phosphorylation in the brain secondary to severe burns.
As a 90 kDa protein, it is not likely that gelsolin easily penetrates into brain to perform its effects. Yet a pioneer study has demonstrated that peripherally expressed plasma gelsolin can affect amyloid-
β dynamics in the CNS in two mouse models of Alzheimer's disease (AD) [
41]. The authors suggested that one possible clearance mechanism might be
via plasma gelsolin entrance into brain parenchyma across the BBB, as reports have indicated that the BBB is compromised in mouse models of AD [
42]. Similarly, an increase in the permeability of the BBB is a common event in thermally injured animals [
8], as also shown in our study by the filling of the lateral ventricles with inflammatory exudates, so it is reasonable to speculate that intravenous infusion of gelsolin could penetrate the BBB into brain parenchyma to attenuate neuroinflammation.
Inflammatory mediators are able to alter cellular metabolism by inducing oxidative stress and mitochondrial dysfunction [
43], resulting in pathologic abnormalities [
44]. Abnormally high levels of cytokines in brain have been found to correlate with both morbidity and mortality in patients with extensive burn injury. In our study, cerebral IL-1β and IL-6 mRNA were up-regulated around 8 h pb and kept increasing throughout the entire period. It is likely that HMGB1 levels were significantly elevated in brain at both 24 and 48 h pb. Gelsolin treatment could significantly reduce expression and release of early as well as late proinflammatory cytokines. This down-regulation of the inflammatory response would lead to less damage and cell loss in the brain, which might, in the future, allow preservation of cognition in patients with severe burn injury. IL-10 is an immuno-suppressant that is mainly secreted by regulatory T cells. It is well known for its positive effects in cerebral ischemia in rats [
45]. We did not detect IL-10 gene expression in brain during the entire observation period, suggesting that administration of gelsolin only inhibits proinflammatory cytokine transcription, without augmenting expression of anti-inflammatory cytokines.
Microglial cells are the primary immune effector cells in brain and play a pivotal role in neuroinflammatory processes associated with a variety of neurological as well as pathological disorders. Microgliosis is a common feature of CNS injury and disease [
38]. Iba-1 is specifically expressed in microglia and plays an important role in regulation of microglial function. Increased Iba-1 immunoreactivity is a hallmark of burn-induced inflammation. It has been proposed that microglial activation induced by sepsis is involved in the pathogenesis of delirium [
46]. There are no reports to date dealing with direct investigation of the activation of microglia and glial scarring following severe burns. We observed kinetic changes in Iba-1
+ cells in striatum and cortex after thermal injury, indicating a highest level of activation as late as 72 h pb. A number of studies have demonstrated that activated glial cells participate in the degeneration of dopamine neurons [
47]. Our data suggest that burn injury
per se might result in microgliosis and loss of vulnerable neuronal populations from inflammation-induced cell death.
Inflammation and apoptosis are two of the most important underlying causes of septic encephalopathy [
48]. Because local accumulation of cytokines may induce apoptosis and significantly extend the initial injury, we also wanted to clarify whether the ability of gelsolin to down-regulate cytokine signaling could lead to decreased activation of apoptotic proteins in brain. Previous investigations have documented that severe burn injury is associated with a significant increase in apoptosis in remote organs [
30,
49,
50] including brain [
47]. A number of markers such as S100B and NSE can serve as general markers of brain injury. Consistent with our observation of morphological improvement, cerebral S100B and NSE levels were diminished by gelsolin infusion. Our study further proves that gelsolin administration immediately following burn injury can reduce caspase-3 activity in brain, confirming a neuroprotective effect of gelsolin.
In inflammatory diseases either inside or outside the CNS, communication between the periphery and the brain
via humoral and/or neural routes results in central neural changes and related behavioral alterations. Monocytes are circulating antigen-presenting leukocytes that play critical roles in inflammation, T-cell differentiation, phagocytosis, and innate immunity [
51,
52]. Previous studies have reported significant infiltration of activated monocytes into the brain of mice with hepatic inflammation [
18], stroke [
19], ischemia-reperfusion [
15] and bacterial meningitis during the post-inflammatory period [
16]. Importantly, these newly recruited monocytes became an integral part of the pool of parenchyma microglia and contribute to the clearance of damaged tissue [
17]. CD11b is expressed by mature monocytes [
16] and monocyte-derived microglia-like cells [
39], whereas CD45 is a pan-leukocyte marker. Resolution of CNS infection is often the result of a balance between immune-mediated pathogen clearance and the deleterious effects of inflammation. Indeed, in a murine model of rabies encephalitis, administration of a sex steroid enhanced permeability of the BBB, promoted immune cell penetration into the CNS, and improved survival [
53]. It has also been reported that gelsolin is necessary for rapid motile responses in cell types involved in stress responses such as hemostasis, inflammation and wound healing [
54]. In gelsolin-mutant mice, macrophage motility was impaired and this contributes to a reduced inflammatory response [
54] and a reduced capacity to recruit macrophages to the injury site, which in turn, slows the clearance of myelin debris and consequently remyelination [
22]. Consistent with these findings, we noticed that gelsolin infusion could accelerate the recruitment of CD11b
+ and CD45
+ cells into the periventricular region of brain early after burn injury, but could still exert a suppressive effect on their recruitment at 72 h pb, indicating an early recruitment of monocyte/macrophage by gelsolin. The increased penetration of CD11b+ cells and the enhanced microglial activation in gelsolin-treated animals were found to be associated with down-regulation of proinflammatory cytokines and caspase-3 activities. Taken together, these results indicate that treatment with gelsolin could ameliorate inflammatory responses in brain and apoptosis of cerebral cells after burn injury.
To elucidate the potential signaling mechanism underlying gelsolin-mediated neuroprotective activity, we examined expression levels of phospho-ERK in burn mice. Western blotting experiments using anti-phospho-p44/42 MAPK (ERK1/2) mouse antibodies revealed activation of phospho-ERK in brain following thermal injury, which is consistent with previous reports [
7,
13]. ERK activation may be downstream of free radicals formation, based on the finding that dopaminergic cells exposed to 6-hydroxydopamine, a reactive oxygen species generating neurotoxin, exhibit a distinct temporal pattern of ERK1/2 activation and caspase-3 activity [
55]. It has been demonstrated that neurons are damaged following prolonged exposure to high concentrations of corticosterone, with activation of p38 MAPK, ERK1/2, and c-jun N-terminal protein kinase 1 [
56], particularly in chronic inflammatory and immune diseases. The increased phospho-ERK levels in brain following burn injury might be a consequence of multiple factors, including proinflammatory mediators, ischemia, and oxidative stress.
We further found that gelsolin treatment dramatically inhibits expression of phospho-ERK1/2 in brain of burned mice. These biochemical results are not in agreement with a previous observation that the neuroprotective effects of estrogen could be attributed to increased phospho-ERK in brain [
13]. However, other authors have reported that administration of neuroprotective reagents reduces phospho-ERK1/2 activity [
57,
58], and inhibition of ERK1/2 can protect against brain damage resulting from focal cerebral ischemia [
59]. Furthermore, it has been demonstrated that gelsolin overexpression inhibits ERK1/2 phosphorylation, nuclear factor of activated T-cell activation, and IL-2 production [
23]. Thus, the exogenous supply of gelsolin in our experiments might protect the brain from exposure to pro-apoptotic stimuli, which in turn might down-regulate ERK1/2 phosphorylation.
Although cerebral complications have been related to increased mortality in severe sepsis [
60] and major burn victims [
10], attenuation of neuroinflammation might not account for all of the benefit of gelsolin in reducing burn-induced mortality in our study. Considering that burn injury could result in severe suppression of the immune system, which plays an important role in the development of subsequent sepsis, multiple organ failure and even death, we examined the dynamic changes in immune function of splenic T cells as well. We found an immunosuppressive effect involving T cells following burn injury, which is consistent with a previous report of perturbed T cell homeostasis after burn injury [
61]. It was encouraging to find that gelsolin infusion could markedly enhance cell-mediated immunity of splenic T cells, which might also contribute to reduced post-burn mortality.
While these studies are intriguing, there are several limitations that should be addressed in future investigations. The first shortcoming is that clinical outcome variables were not obtained. For clinical relevance, multi-organ dysfunction which may be the root cause of burn-induced mortality should be evaluated in further studies. Secondly, neurological outcomes like edema, BBB penetrability and cognitive function were not assessed in the current study. Better understanding of the improvement of neurological outcome with gelsolin may allow an in-depth understanding of the mechanisms by which gelsolin attenuates the acute response, and to what extent neurological damage contributes to post-burn mortality. Although we initiated this study to observe the acute effects of gelsolin on neuroinflammation following burn injury, it may be possible to solidify our current observations in a further study by also evaluating the effects of gelsolin on these neurological complications which are frequently seen in our clinical patients. Thirdly, with regard to the time-window of gelsolin delivery, intravenous infusion of gelsolin immediately after burn injury resulted in significantly reduced mortality. However, the interventional time could be postponed to later intervals to more closely simulate a clinical setting for this therapeutic strategy. The final, but foremost concern regards the pharmacokinetics of gelsolin in this model. With a half life as long as 2.3 days [
62], a single administration of gelsolin could produce considerably elevated gelsolin levels as early as 8 h and remained high at 72 h pb. As BBB disruption may occur as early as 7 h after burn injury [
8], while gelsolin might not penetrate the BBB directly within the first hours, and it is reasonable to speculate that gelsolin could breach the BBB to perform its effect directly in the brain at later time points. Nevertheless, the precise mechanism of our observed gelsolin effect on response to thermal injury and immunomodulation in both the brain and the periphery requires further studies.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
QHZ participated in the design of the study; personally conducted a significant portion of the experiments presented in the manuscript, and participated in the writing of the manuscript. QC participated in the design of the study and the preparation of the animal model. JRK prepared all the cryostat sections. LC and XMZ did the cell counting of the brain. ND conducted the QPCR detection. ZYS supervised and edited the manuscript. YMY participated in the design of the experiments, funding of the projects, and preparation of the manuscript. All authors have read and approved the final version of the manuscript.