Background
Traumatic brain injury (TBI) is a major health concern in contact sports and in the military and is the leading cause of death and disability in children and young adults. Though progress has been made in prevention and acute clinical management, a therapeutic that ameliorates the disability or mortality associated with TBI remains elusive. Despite the growing awareness of this public health problem, a greater translational effort in clinical trials for TBI is needed [
1]. Expanding the preclinical research landscape might enhance our understanding of TBI pathophysiology and facilitate the discovery of therapeutics that improve recovery outcomes.
Secondary damage in surrounding brain tissue occurs following the primary insult. Pro-inflammatory mediator expression is enhanced soon after injury and is followed several days later by a period of enhanced expression of anti-inflammatory cytokines [
2]. While the immune response in the days and weeks after TBI can benefit recovery by clearing cellular debris and producing neurotrophic factors, disproportionate expression of neurotoxic pro-inflammatory mediators in the first hours and days after TBI injury may be detrimental [
3]. Thus, an excessive and/or persistent innate immune response emanating from the primary injury site may contribute to the severity of secondary damage.
The time course of inflammatory mediator expression is well-established in
in vivo rodent models of TBI. Interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α) increase within the first several hours after cortical injury but then return to baseline levels by 24 h [
4-
6]. Chemokine ligands 2 (CCL2, formerly known as monocyte chemoattractant protein-1), 3 (CCL3), 4 (CCL4), and prostaglandin-endoperoxide synthase 2 (PTGS2) increase early after injury and remain elevated for 24 h or longer [
7-
9]. IL-1β and TNF-α are cytokines associated with promoting further inflammation, vascular permeability, and neurotoxicity. Chemokines recruit inflammatory cells and enhance synthesis of other inflammatory mediators. PTGS2 converts arachidonic acid from the damaged cell membranes to prostaglandins that modulate local cerebral blood flow, recruit inflammatory cells, and enhance free radical synthesis. In these rodent models, enhancing cytokine levels or genetic removal alters post-TBI outcomes highlighting the importance of these mediators in recovery early after TBI [
6,
8,
10].
While most of our understanding of the early expression and effects of inflammatory mediators in response to cortical injury is from the rodent models of TBI, the timing and contribution of long-term deficits in larger animal models and in human patients are less clear. An increase in TNF-α and IL-1β is detected in human cortical tissues early after injury [
11]. An increase in CCL2 in human CSF after TBI is found early and remains elevated for over 12 h [
8]. CCL3 and CCL4 in cerebral microdialysis fluid are also elevated for 24 h or longer [
12]. It remains unclear whether regulating the synthesis of these pro-inflammatory mediators will be helpful. Treating human TBI with synthetic glucocorticoids, which inhibit the synthesis of many inflammatory mediators, has not provided favorable results [
13]. Safer anti-inflammatory alternatives to glucocorticoids are being investigated and show promise [
14].
Discovering new therapeutic agents with ideal immunomodulatory characteristics in the CNS may improve outcomes when administered early after TBI. This discovery effort will be facilitated by filling the gaps in preclinical translational research that exist between cell culture and rodent models and between rodent models and human studies. To fill these two gaps, we examine the early changes in the mRNA expression of inflammatory mediators following cortical injury in two understudied systems of TBI. In the first system, we use mouse cortical slices to evaluate the
ex vivo changes in the mRNA expression of cytokines, chemokines, and PTGS2 that occur during the first 24 h after injury. Slices maintain neural cells in their normal microenvironment, and an acute innate immune response occurs following tissue sectioning. In the second system, we used piglets to assess the
in vivo changes in the expression of these inflammatory genes 24 h after cortical impact at the lesion core and in the penumbra region. The piglet brain is gyrencephalic where the abundance of white matter, anatomy, vasculature, and pattern of development closely parallels humans [
15]. These two systems of cortical injury both demonstrate increased expression of genes involved in the innate immunological response.
Discussion
The immunological response to neural injury is predominantly investigated in the
in vivo rodent models. Alternatively, glia are activated
in vitro with endotoxins or cytokines in dissociated cell culture systems. Yet, therapies identified in current model systems have not easily translated to clinically effective treatments [
23]. Additional preclinical systems will expand our understanding of the acute inflammatory response to cortical injury and may facilitate the discovery of pharmacological agents that attenuate secondary damage. In this study, we report the transcriptional changes of inflammatory genes in two cortical injury systems that can help fill the preclinical translational gaps in knowledge that exist before and after
in vivo rodent models.
Microglia are the resident CNS cells derived from monocyte/macrophage lineage and are the early responders to neural injuries. While microglia can be isolated or immortalized for use in highly valuable cell culture studies, one disadvantage to this system is that dissociated microglia are separated from their normal neural contacts. Microglial phenotype and function are regulated by the unique and complex cellular and extracellular signaling molecules [
24]. Studying slices of CNS tissue
ex vivo has the advantage of maintaining glia in their neural microenvironment, and the sectioning of tissue causes an immunological response without the requirement of endotoxins. While the change in acute expression of inflammatory mediators in spinal cord slices reveals a similar profile to observations in spinal cord injury in the rodents
in vivo [
25,
26], it is not known whether slices of cerebral cortex will demonstrate a similar pattern to
in vivo models of TBI. The IL-1β and TNF-α expressions increase in the cortex within the first few hours in the rodent models of TBI before returning to baseline levels by 24 h [
4,
5]. Similarly, the expression of these cytokines increased in the first 6 h and then decreased by 24 h in mouse cortical slices. In the
in vivo rodent models, the expression of chemokines and PTGS2 increases within hours of cortical injury and, unlike IL-1β and TNF-α, remains at elevated levels at 24 h or beyond [
7-
9]. Remarkably, these particular transcripts increased early and remained elevated at 24 h in cortical slices. These findings suggest that cells resident to the CNS are largely responsible for the acute immunological response to injury because the
ex vivo slice system has negligible infiltration of leukocytes. The inflammatory gene expression in slices is regulated by a genetic background [
16] and can be modified by pharmacological agents in the explant medium [
25-
27]. The slice system of cortical injury represents an uncomplicated method to investigate the early immunological response and is amendable to transgenic experiments and for screening of active anti-inflammatory agents that target inflammatory expression at the level of transcription. Future studies should examine whether expression of cytokines and chemokines is increased at the protein level, which will enhance the utility of the cortical slice injury system.
While cortical slices help fill the translational gap between cell culture and intact rodent models of injury, a gap in knowledge of acute cortical injury response also exists between
in vivo rodent models and clinical studies in humans. Swine have gyrencephalic brains similar to humans in which 50% of the brain is comprised of white matter in contrast to the lissencephalic rodent brain that contains 15% white matter. Outbred swine, which are less expensive than primate species and physiologically similar to humans, are increasingly being used to study other trauma and other diseases involving the immune system such as organ transplant biology as they have fewer immunologic hurdles to overcome from swine to human [
28]. Swine are increasingly being utilized as models for pediatric traumatic brain injury [
18,
29] and blast injuries in the military [
30]. Swine are particularly useful in understanding TBI in children as the timing of the peak growth spurt in the piglet occurs around the time of birth in humans but occurs prior to birth in primates [
15]. TBI superimposed upon development adds complexity to the problem of therapeutics as therapies may have the opposite effect in the immature brain compared to the mature brain and may interfere with normal development to a greater degree of any benefit in treating the brain injury [
29].
This report demonstrates the early transcriptional inflammatory response following cortical injury in piglets. In agreement with previous findings using outbred swine [
22], which parallel genetic diversity in humans, we found a high degree of variability in mRNA expression for some genes. PTGS2 mRNA in the cerebral cortex was particularly inconsistent in piglets. IL-1β and TNF-α mRNA levels were also variable in the lesion core 24 h after injury. Since these cytokines increase in the early hours after injury in the rodent models of TBI before returning to baseline at 24 h [
4,
5], future studies should evaluate IL-1β and TNF-α expressions at earlier time periods after injury in piglets. Chemokines, on the other hand, were more stable and represent reliable markers of acute inflammatory changes in response to injury in piglets. Increased CCL2 mRNA was found in the lesion core 24 h after injury. An increase in CCL2 was also found in the penumbra region, which revealed a positive linear correlation to changes in the lesion core within piglets. CCL2, as opposed to PTGS2, may represent a more reliable biomarker for injury severity and for drug discovery efforts aimed at regulating the immunological reaction. Insight into stable and inconsistent inflammatory transcripts in response to TBI in outbred swine may help in identifying reliable biomarkers useful in the development of therapies to overcome the inherent variability in humans. Here, we initiate study into the inflammatory response in immature swine and plan on determining the age-dependent differences in the acute injury response, which may explain the previously described age-dependent lesion size among stages of immaturity and reveal age-specific inflammatory targets [
18,
19]. Here, we have used injured piglets as their own control comparing cortical tissue from the ipsilateral vs. contralateral tissue similar to our previous studies [
18,
29]; however, immune modulators may indeed be elevated in the contralateral hemisphere thus blunting our observed increases and perhaps contributing to the variability among subjects in the contralateral cortical tissue. Future studies should include sham piglets to determine if the inflammatory modulators are increased in the cortex contralateral to the impact in this immature, large animal model of TBI.
New experimental systems to investigate the acute immunological reaction may facilitate the discovery of early interventional drugs to mitigate the disability and fatality after TBI. In the current study, two disparate mammalian systems of cortical injury illustrate an early increase in pro-inflammatory transcripts. The mouse cortical slice system reveals a temporal change in cytokine and chemokine expressions that is similar to the in vivo changes reported with the rodent models of TBI. This slice injury approach can bridge the gap in experimental systems that exists between higher throughput cell culture and lower throughput animal models to facilitate immunotherapeutic discovery and the understanding of the acute immunological reaction to cortical injury. Later stages of preclinical development of immunomodulatory drugs would benefit from injury models using swine, which may more reliably model the brain anatomy and subject variability in human studies. The ex vivo mouse cortical slice injury system and the in vivo piglet model of brain injury can serve as additional preclinical systems to facilitate the discovery of therapeutic agents for TBI aimed at regulating early inflammatory mediator expression.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
DJG performed the mouse slice experiments. BAC performed the piglet experiments. DJG, BAC, and WFH contributed to the study design, drafted the manuscript, and approved the final manuscript.