Lack of chronic neuroinflammation in the absence of focal hemorrhage in a rat model of low-energy blast-induced TBI

Military personnel exposed to blast overpressures are at risk of developing behavioral and cognitive abnormalities [11]. Acutely, several studies in animal models have shown that blast exposure induces brain inflammation and increased levels of pro-inflammatory factors. Infiltration of polymorphonuclear leukocytes and lymphocytes is observed in the brain parenchyma within 1 h post-blast exposure [53]. In rats, a single 551-kPa blast exposure induced glial fibrillary acidic protein (GFAP), S100β (an astrocytic marker), cyclooxygenase (COX)-2, interleukin (IL)-1β, and tumor necrosis factor (TNF)-α, and these changes were present by 1 h and remained detectable at 3 weeks post-injury [53]. Similarly, GFAP levels were significantly higher in all measured brain regions of rats exposed to 133.8-kPa blast overpressure [26]. In another study, a 120-kPa blast triggered release of IL-1β, TNF-α, IL-10, and erythropoietin (EPO) as early as 3 h after injury, and peak levels were reached at 24 h before they diminished by 48 h [33]. Moreover, a 129-kPa blast induced transcriptional upregulation of the proinflammatory interferon (IFN)-γ and monocyte chemoattractant protein (MCP)-1 by 4 h post-blast, and increased levels of these proteins were detected 24 h post-blast while strong Iba1-immunoreactive microglial cell activation was detected at 2 weeks post-blast [8]. Elevated levels of IFN-γ and IL-6 have also been found in the amygdala and ventral hippocampus, together with an increased number of Iba1-expressing activated microglia following blast injury [26].

A cDNA microarray analysis of the brains of mice exposed to a 142-kPa blast showed significant upregulation of several interleukin receptors in the midbrain [55]. In the frontal cortex, hippocampus, and cerebellum, the expression of IL receptors was significantly reduced, whereas the expression of TNF-α and its receptors was increased [55]. In rats subjected to multiple 138-kPa blast exposures (5×), plasma levels of vascular endothelial growth factor (VEGF) and neuron-specific enolase (NSE) were significantly elevated after 2 h compared to levels in sham controls and rats exposed to a single blast [27]. By day 22 post-injury, animals exposed to either a single blast or multiple blasts had significantly higher levels of VEGF, neuron-specific enolase (NSE), neurofilament H (NFH), and GFAP than did those in the non-injured control groups [27]. At 1 day post-blast exposure, increased GFAP immunoreactivity was observed in the hippocampus of animals exposed to a single blast, whereas no such increase was seen in animals exposed to multiple blasts [27]. However, by day 22, an apparent increase in GFAP immunoreactivity was observed in the hippocampus of the multiple blast-exposed rats. In both the single and multiple blast-exposed groups, increased apoptosis was found in the hippocampal hilus [27] .

In a temporal evaluation of cytokines in rat serum after a single 117-kPa blast exposure, a significant decrease in IL-1α expression was observed at 3 h as well as a decrease in M-CSF expression at 24 h, an increase in EPO at 48 h, and decreased levels of IL-1α, IL-1β, IL-6, IL-10, and EPO along with increased levels of VEGF and macrophage colony stimulating factor (M-CSF) at 72 h post-blast [44]. Blast injury (142 kPa) in stressed rats resulted in elevated levels of serum corticosterone, NFH, NSE, GFAP, and VEGF compared to levels in non-blast-exposed controls 2 months post-injury. The hippocampus and prefrontal cortex (PFC) also contained increased numbers of apoptotic cells (TUNEL-positive) and elevated levels of GFAP, S100β, Iba1, VEGF, IL-6, IFN-γ, and phosphorylated tau [32]. In humans, 35- to 81-kPa blast exposures result in increased levels of IL-6 and TNF-α in the serum that were maintained for 24 h [18].

These data altogether suggest that acute blast exposure induces brain inflammation and increased levels of pro-inflammatory factors. Acute inflammation has long been considered a transient phenomenon in many forms of TBI. However, there is accumulating evidence that the inflammatory response after TBI may persist [14]. Indeed, studies in animals document persistent inflammation in brain after TBI [14]. Postmortem human studies find inflammatory changes that can persist for years after TBI [24], and a recent study using positron emission tomography to image a ligand that targets microglia found increased microglial activation up to 17 years after injury [43] .

We previously reported that rats exposed to low-level blast overpressures (74.5 kPa) develop a post-traumatic stress disorder (PTSD)-like phenotype associated with chronic brain vascular degeneration and rarely hemorrhagic brain cortical tears that follow the lines of penetrating vessels [17, 49]. In the present study, we investigated whether three 74.5-kPa blast exposures can induce chronic microglial activation and increase the levels of pro-inflammatory factors in brain. Our results show that under this blast protocol and in the absence of vascular disruption, low-level blast exposures do not alter the microglial cell density nor microglial activation in the hippocampus and prelimbic cortex and do not significantly alter the levels of cytokines involved in neuroinflammation (over 6–40 weeks post-blast). However, in the presence of focal hemorrhage, neuroinflammation with increased microglial activation and astrocytosis was observed 16 weeks post-blast exposure. These results show that neuroinflammation is induced by leakage of blood elements from a fragile vasculature into the brain parenchyma. However, in the absence of hemorrhage, chronic neuroinflammation is not a general feature of low-level blast injury.

Acute inflammation is a recognized feature following TBI induced by mechanical trauma [47] with recent evidence that a chronic neuroinflammatory component may also develop and contribute to the late effects of TBI including the development of neurodegnerative diseases [14]. Inflammation is also a component of acute blast injury [12]. Activation of resident microglia, influx of peripheral leukocytes, and release of proinflammatory mediators are early events of this process. Microglia are activated by thrombin through MAPK signaling pathways via the proteinase-activated receptor-1 (PAR-1) [16, 39]. Microglia can also be activated through Toll-like receptors (TLRs) and the receptor for advanced glycosylation endproducts via danger-associated molecular signals (ATP, neurotransmitters, nucleic acids, heat shock proteins, and high-mobility group box 1 proteins) released by necrotic cells [10, 40, 52]. Microglial activation leads to increased production of TNF-α and IL-1β, which induces neuronal apoptosis [57]. Microglia are also involved in clearance of cell debris and phagocytosis of blood components, with a central role in hematoma resolution.

Explosive blasts rapidly generate very high levels of kinetic energy that dissipate as supersonic pressure waves to cause blast-induced TBI [13, 37, 42]. In the brain parenchyma, these high-energy, high-velocity blast waves can also cause substantial damage to blood vessels as well as to neuronal and glial cell bodies and their processes [4–6, 17, 19, 28, 49]. Prolonged but not short-duration high-energy blast waves (620–1570 kPa) result in the acute onset of neuroinflammation and of increased levels of pro-inflammatory cytokines in the brain [9]. Depending on the intensity of the blast, TBI may include an early-onset diffuse cerebral edema and delayed vasoconstriction [3, 34–36]. Injury secondary to blast-induced TBI involves vascular remodeling, neuroinflammation, and gliosis that are visible several months after the initial injury [6, 28, 37, 51].

In contrast to these findings after high-energy blast exposures, our experiments with lower level energy blast exposures (74.5 kPa) did not demonstrate the presence of chronic neuroinflammation 6 weeks post-blast exposure. Immunohistochemical analyses of brains from blast-exposed animals without any evidence of vascular leakage did not show obvious microgliosis, as shown by the relatively low abundance of Iba1⁺ reactive or amoeboid microglia (types 3 and 4) expressing MHCII, and did not present major alterations in the brain inflammasome even at 40 weeks post-blast exposure. Curiously, lack of inflammation after mild brain injury has also been reported in a mouse model of closed head injury using a standardized weight-drop technique [45]. The lack of inflammation observed in our animals indicates that low-energy blast exposures (74.5 kPa) are not always sufficient to sustain chronic neuroinflammation. In a murine model system, microglial activation associated with microdomains of vascular disruption (tight junction injury) has been observed 45 min post 105.5-kPa blast exposure [22]. However, by 14 days post-blast, elevated levels of TNF-α were only sustained in animals exposed to three repetitive blasts, suggesting that even at higher blast energy, repetitive exposures are required to promote more persistent neuroinflammatory changes in the CNS [22].

In blast-induced TBI, vascular blood leakage may be a requirement for the progression and persistence of neuroinflammation. This is also supported by an observation in a blast-exposed animal that exhibited an infiltrated clot within scarred tissue 16 weeks post-blast exposure that was associated with overwhelming microglial activation (Figs. 6 and 7). As the clot was found 16 weeks post-blast exposure, it could be presumed that the vascular leakage occurred at a much later time after the last blast exposure, most likely as a result of a progressive blast-induced vascular degenerative processes, leading to induced vascular fragility, subsequent rupture, and blood leakage [17]. We have previously reported the selective vulnerability of penetrating vessels that follow the patterns of blast-induced focal tears and of the associated microvasculature to 74.5-kPa blast exposures. Vascular injuries are present acutely at 24 h after blast exposure and lead to a chronic vascular degenerative processes that can lead to vascular rupture later in life [17].

Comparison of high- and lower-energy blast-induced TBI clearly indicates that BBB leakage, microglial activation, and increased production of pro-inflammatory cytokines depend on the blast overpressure energy and the number of exposures. Acutely, after 72 h, animals exposed to 74.5-kPa blasts present pathological alterations that mainly included blood leakage from the choroid plexus into the lateral ventricles, focal non-hemorrhagic tissue tears, and vascular alterations [17, 49]. In other rat models, blast-induced BBB leakage appeared preferentially at higher blast pressures (>110-kPa), as it was shown that extensive leakage occurred in all brain regions but preferentially in the thalamus, striatum, hippocampus, and occipital cortex [25, 29]. However, limited leakage, mainly through the chorionic plexus, was observed after exposure to 72-kPa blasts [25, 29, 49]. Kawoos et al. [29] also showed that there is a qualitative relationship between BBB leakage and increased intracerebral pressure (ICP), through which the ICP levels and sustainability depend also on blast intensity and the number of blast exposures. As blast induces an early vasodilation (as evidenced by enlarged blood vessel diameters [25]), the intravascular forces exercised by the pressurized circulating blood could damage the vascular smooth muscle and endothelial layers [2, 17, 19, 20, 25, 29, 49, 56]. Through buckling, it could induce vascular tortuosity [21] and through subsequent axial stretch [15], generate vascular strictures [17, 49]. In small vessels, endothelial damage can also lead to a breakdown of the BBB and blood leakage [25, 29]. In large vessels, endothelial damage could trigger chronic vascular disease through vascular remodeling by induction of neointima formation, neointima thickening, restenosis, aneurysm formation, plaque deposition, vascular occlusion, thromboembolism, and vascular rupture. Therapeutic approaches aimed at preventing or reversing vascular damage may improve the chronic neuropsychiatric symptoms associated with blast-induced TBI.

Recently much interest has been generated in the role of ongoing neuroinflammation in the chronic effects of TBI. We investigated whether chronic neuroinflammation was present in a rat model of repetitive low-energy blast exposure. No significant alterations in the levels of neuroinflammatory indicators (microglia proliferation and activation as well as pro-inflammatory cytokine/chemokine concentrations) were observed in rats 6-40 weeks after three 74.5-kPa blast exposures. Microgliosis and microglial activation were observed in one animal associated with vascular blood leakage 16 weeks post-blast, most likely due to chronic vascular degeneration. Thus, extravasation of blood elements may be a trigger for neuroinflammation in blast-induced TBI. However in the absence of hemorrhage, chronic neuroinflammation does not appear to be a long-term consequence of low-level blast injury.