Src Family Kinases in Brain Edema after Acute Brain Injury

Abstract

Brain edema, the first stage of intracranial hypertension, has been associated with poor prognosis and increased mortality after acute brain injury, such as ischemic stroke, intracranial hemorrhage (ICH), and traumatic brain injury (TBI). The acute brain injury often initiates release of many molecules, including glutamate, adenosine, thrombin, oxyhemoglobin, cytokines, reactive oxygen species (ROS), damage associated molecular pattern molecules (DAMPs), and others. Most of those molecules activate Src family kinases (SFKs), a family of proto-oncogenic non-receptor tyrosine kinases, resulting in blood-brain barrier (BBB) disruption and brain edema at the acute stage after brain injury. However, SFKs also contributes to BBB self-repair and brain edema resolution in the chronic stage that follows brain injury. In this review we summarize possible pathways through which SFKs are implicated in both brain edema formation and its eventual resolution.

Introduction

Brain edema occurs when a cerebral blood vessel is blocked or ruptured following ischemic stroke, intracerebral hemorrhage (ICH), traumatic brain injury (TBI) and other neurological diseases [1–3]. There are two main categories of brain edema, namely cytotoxic (cellular) edema and vasogenic (extracellular) edema [4]. In cytotoxic edema, the blood-brain barrier (BBB) remains intact, but there is essentially a compartment shift of water from the extracellular to the intracellular compartment, with no increase of brain water content or rise in ICP. Though it does not require BBB disruption, cytotoxic brain edema changes cellular metabolism and eventually damages BBB after brain injury. By contrast, vasogenic edema requires BBB disruption, allowing fluid (i.e., circulating blood) to accumulate in the extracellular space in brain parenchyma and will increase ICP [4]. It is generally thought that cytotoxic edema is dominant immediately following ischemic stroke [5], while vasogenic edema is dominant at the acute stage after TBI [4]. However, cytotoxic and vasogenic edema usually combine when brain injury progresses into the chronic phase in which a characteristic breakdown of BBB occurs no matter what type of edema was first in the acute stage post brain injury [6]. Therefore, maintenance of BBB integrity has become a focus of recent research to prevent brain edema and improve outcomes of acute brain injury.

Brain edema has been associated with high mortality, mostly because it can induce rapid increase in intracranial pressure (ICP), which leads to compression of blood vessels, reduced tissue blood flow, reduced oxygenation and shifts tissue down pressure gradients (herniations) that may crush vital brain centers and eventually cause respiratory or heart failure [4]. An aggressive treatment for raised ICP can reduce mortality and improve outcome [7, 8], though ICP control alone (i.e. osmotherapy) may be insufficient to benefit long-term recovery after brain injury [9]. This is probably because osmotherapy is unable to block the release of many toxic molecules that follow acute brain injury, such as glutamate, adenosine, oxyhemoglobin, thrombin, cytokines, reactive oxygen species (ROS), damage associated molecular pattern molecules (DAMPs) and others [10–40]. These molecules mediate BBB disruption and brain edema through multiple ligand-receptor pathways. Since brain edema might occur via many parallel pathways, blocking just one or two of these pathways may not be clinically effective in treating human brain injury [16].

Src family kinases (SFKs), a family of proto-oncogenic, non-receptor tyrosine kinases, include nine family members: c-Src, Fyn, Yes, Yrk, Lyn, Fgr, Hck, Blk and Lck [41–43]. They can be activated by many trans-membrane receptors, such as adhesion receptors, tyrosine kinase receptors, G protein-coupled receptors, cytokine receptors, and others [44]. This makes SFKs a point of convergence for many molecules, and targeting SFKs has potential to prevent disruption of BBB components (i.e., endothelial cells, astrocytes, pericytes, neurons, tight junctions, and others) and block brain edema via modulating their multiple downstream targets, such as NMDA receptors [45–50], mitogen-activated protein kinases (MAPKs) [51–57], and cyclin-dependent kinases (Cdks) [58–62]. Many studies have demonstrated that acute administration of SFK inhibitors (e.g., PP1, PP2) attenuates BBB breakdown and prevents brain edema after acute brain injury [18–20, 63–66]. However, delayed and chronic administration of PP2 prevents the BBB self-repair and lengthens the period to resolve the edema in the recovery stage after brain injury [20]. These suggest SFKs may play dual roles in both brain edema formation and resolution during the different stages following acute brain injury (Figure 1).

Figure 1.

Activation of SFKs results in BBB disruption and brain edema formation in the acute stage, but leads to BBB self-repair and brain edema resolution in the recovery stage after acute brain injury, such as ICH, TBI and ischemic stroke.

Tissue specificity, structure, activity and functions of SFKs

Several SFK family members (c-Src, Fyn, Yes, Yrk) are ubiquitously expressed, whereas others (Lyn, Fgr, Hck, Blk, Lck) are generally found in brain and hematopoietic cells [47, 67–72]. In adult mice, Fyn and c-Src mRNA expression is highest in hippocampal neurons [73, 74]. Importantly, one tissue can express multiple SFK members, for example, Src, Fyn, Yes, and Lck have been examined in brain [47, 67–72], and the different SFK family members are often found to compensate for one another [75].

Structurally, SFK family members share a conserved domain structure consisting of consecutive SH3 (polyproline type II helix for protein-protein interaction), SH2 (phosphotyrosine recognition), and SH1 (tyrosine kinase catalytic activity) [43]. They also contain a membrane-targeting region at their N-terminus that is followed by a unique domain of 50–70 residues, which is divergent among family members [43]. Although it still incompletely clear, Src activity is regulated by tyrosine phosphorylation at two sites (one is at Tyr416 in the SH1 domain, the other at Tyr527 in the short C-terminal tail), but with opposing effects. While phosphorylation at Tyr416 activates Src, phosphorylation at Tyr527 results in inactivation [72, 76].

Under normal physiological conditions, SFKs are implicated in the regulation of embryonic development, cell growth, cellular differentiation and inflammatory responses [74, 77–79]. SFKs can initiate negative feedback to prevent their sustained activation via recruitment of inhibitory factor C-terminal Src (Csk) [80]. The feedback loop consists of SFK activation leading to phosphorylation of Csk binding protein (Cbp), and the phosphorylated Cbp targets Csk to SFKs and promotes inhibitory Csk phosphorylation of SFKs [81].

Due to mutations in SFKs or Csk, aberrant activation of SFKs can occur in cancers, and the abnormal SFK signaling contributes to many aspects of tumor development, including proliferation, survival, adhesion, migration, invasion, as well as metastasis [82–84]. Thus, it is likely that targeting SFKs may be a promising therapeutic approach for cancer, as SFK antagonists have been tested and well tolerated in cancer clinical trials [85–88].

Recently, we and others have demonstrated a new function of SFKs in acute brain injury, that is, transient activation of SFKs associated with BBB disruption, brain edema and spatial memory deficits following experimental ICH (intracerebroventricular fresh blood or thrombin model), TBI (lateral fluid percussion (LFP) model), and stroke (middle cerebral artery occlusion (MCAO) model) [18, 19, 63–66].

SFK activation, excitotoxicity, BBB disruption and brain edema

Following acute brain injury (i.e., ICH, TBI, ischemic stroke), there occurs a transient increase of glucose utilization and local cerebral blood flow [53, 89, 90], presumably because of the actions of glutamate in blood at the time of brain injury. This was supported by findings that glucose hypermetabolism could be blocked by antagonists of NMDA and AMPA receptors [53, 90]. However, glutamate alone could not explain the hypermetabolism since glutamate injected directly into brain does not produce hypermetabolism [53]. This suggests that acute brain injury affects NMDA receptors in some way to make them more sensitive to glutamate in order to mediate brain injury and/or hypermetabolism.

A large number of studies have revealed that the molecules released following acute brain injury (e.g. adenosine, thrombin, cytokines) can activate SFKs [10–32], and SFKs directly bind NMDA receptors and modulate their activity [45–50]. Our data show that either an NMDA receptor inhibitor (MK801) or an SFK inhibitor (PP2) is able to prevent brain edema and improve behavioral outcomes after intracerebroventricular injection of thrombin in rats [19]. Therefore, it is plausible that SFKs and NMDA receptors are coupling to mediate calcium overload, glucose hypermetabolism and brain edema after acute brain injury.

SFK activation, mitogenic signaling, brain edema formation and resolution

SFKs can be activated by many trans-membrane receptors, such as adhesion receptors, tyrosine kinase receptors, G protein-coupled receptors, cytokine receptors, and others [44]. This unique feature of SFKs makes them a point of convergence for many toxic molecules that are released after brain injury [10–40]. Most of those molecules are abruptly released and reach peak concentrations within a couple of hours to a day after brain injury. In the acute stage over-activated SFK mitogenic signaling causes neurons to enter the cell cycle and die, and damages astroctyes and endothelial cells via MAPKs or CdKs [14, 19, 20, 51–62]. The disruption of BBB components increase BBB permeability, resulting in brain edema after acute brain injury.

Within about a day after acute brain injury, the molecules resolve gradually, and the disease progresses to a recovery stage of brain injury. The restored moderate SFK/mitogenic signaling leads to birth of new endothelial cells, astrocytes and other cells that mediate BBB self-repair and brain edema resolution. Recent studies suggest that a number of stem cells exist throughout the mammalian brain, and some of these are associated with vascular niches [91]. Such stem cells could serve as a source of newborn endothelial cells, astrocytes and other cells of the neurovascular unit that would play a major role in re-establishing the BBB after brain injury [92].

Though SFK inhibitors prevent toxicity signaling at the acute phase after ICH, they also block cellular proliferation of stem cells to delay and prolong BBB self-repair [55, 56, 93, 94]. This may provide at least a partial explanation for the findings that: (1) acute single administration of SFK inhibitors (PP2, 1mg/kg, i.p. immediately after ICH) can attenuate the intracerebroventricular injection (i.c.v.) of thrombin-induced BBB disruption and brain edema [20, 52]; (2) and that delayed and chronic administration of SFK inhibitor (PP2, 1 mg/kg, i.p. daily, day 2 through 6) prevents thrombin-induced BBB repair and brain edema resolution in rats [20, 52].

Additionally, SFKs also activate hypoxia-inducible factors (HIFs) that can increase BBB permeability for brain edema formation or promote angiogenesis for brain edema resolution after brain injury through expression of aquaporins (AQPs), matrix metalloproteinases (MMPs), vascular endothelial growth factor (VEGF), BBB proteins (i.e., occluding), and others [95–97]. Interactions and cross-talk with these and other molecules and pathways add complexity to timing and development of appropriate treatment strategies involving the SFKs.

Future directions

Future studies need to address exactly which specific SFK members found in brain (e.g., Src, Fyn, Lck and Yrk) mediate edema following acute brain injury. In view that SFKs also play critical roles in brain edema resolution, the therapeutic time window of SFK inhibition should be studied for treating edema following acute brain injury and avoid the potential side effects caused by chronic inhibition of SFKs. A nanoparticle-based siRNA transfection system can be used for knockdown of individual SFK genes, as it allows transient knockdown of target genes, high efficiency of in vivo siRNA delivery, high specificity for gene targets, low cytotoxicity [98, 99], and is approved by the FDA for human use [100–103].