Background
Neuroinflammation and degeneration occurs following hypoxic-ischemic insults such as traumatic brain injury (TBI) or chemical exposure to neurotoxic agents [
1]. Neuroinflammation and degeneration often share common pathways frequently leading to neuronal cell death [
2]. Complement represents an important mediator during the neurodegenerative process by releasing proinflammatory mediators and anaphylatoxins such as C3a and C5a as well as producing MAC [
3]. Complement fragments and C3aR have been demonstrated in normal and ischemic brain tissue [
4]. Complement depletion has been shown to reduce post-ischemic brain injury in rats and mice [
4]. It has been suggested that complement activation levels in the central nervous system (CNS) following brain injury might increase after blood brain barrier (BBB) break-down [
5,
6] and might come from cellular sources such as astrocytes, microglia, oligodendrocytes and neurons in response to cerebral ischemia or brain trauma [
1,
7,
8]. In addition, astrocytes and microglia express complement inhibitors on their membranes to control complement activation and mitigate complement-mediated injury [
9]. Neurons express low levels of complement regulators compared to astrocytes and it has been suggested that human fetal neurons have the capacity to spontaneously activate the complement system [
10].
Inhibition of complement activation using biologics such as soluble complement receptor type 1 (sCR1), C1 inhibitor (C1-INH), C3 convertase inhibitor (Crry), C5a monoclonal antibodies, and C5a receptor antagonists have been shown to reduce post TBI [
4,
11]. Complement system can be activated via the classical pathway, such as by IgG activation, or by the alternative pathway, such as by factor B activation [
12]. In a recent study, intravenous immunoglobulin (IVIG) was demonstrated to protect the brain against injury from experimental stroke in mice [
4]. Therefore, targeting the complement cascade represents a potential treatment strategy for the management of ischemic brain injury.
Decay-accelerating factor (DAF, also known as CD55), a ubiquitously expressed intrinsic complement regulatory protein, inhibits complement activation by inhibiting the function of C3/C5 convertases in both the classic and alternative pathways thereby limiting local C3a/C5a and MAC production [
13]. Human NT2-N neurons constitutively express DAF which is down-regulated after severe hypoxia and subsequent reoxygenation with human serum [
14]. It has been previously shown that increased expression of DAF plays an important role in the reduction of cerebral damage by steroids after Traumatic Brain Injury [
15]. It has been demonstrated that inhibition of complement activation by human recombinant DAF results in local and remote tissue protection during mesenteric ischemia/reperfusion in mice [
16]. A common model of chemical ischemia in cultured cells involves exposure to cyanide [
17]. In the present study, we evaluated the effect of recombinant human DAF on cultured embryonic rat primary neurons subjected to chemically induced hypoxia. Harris et al. 2000 reported that neither human nor rodent DAF are species restricted meaning they can regulate both homologous and heterologous complement activation, suggesting cross-reactivity between human recombinant DAF in rodent preparations [
18]. Results indicate that 200 ng/ml of DAF treatment protects rat neurons from injury by suppressing the complement cascade as well as by inhibiting the activation of caspase and Src tyrosine kinase.
Discussion
Two major questions have been addressed in this study. First, does the presence of soluble DAF attenuate neuronal damage induced by the chemical hypoxic conditions? Second, how does DAF protect neurons from the hypoxia-mediated injury? The results demonstrate that treatment with DAF protects cellular function and increases cell viability in a cultured neuronal chemical hypoxia model. DAF decreases complement activation and distribution. Furthermore, this protective effect of DAF is associated with the ability to directly inhibit complement activation and suppress (directly or indirectly) Src tyrosine kinase and caspase signaling pathways.
Cyanide is a toxic chemical that has been used as a weapon of war and also as means of terrorist attacks on civilian populations [
25]. It inhibits the respiratory chain, blocking the utilization of oxygen and affecting mitochondrial function [
17]. Neurons are particularly vulnerable to energy deprivation [
26]. Therefore, one of the main target organ system of cyanide toxicity is the central nervous system. NaCN treatment, which mimics acute hypoxic cell damage, is commonly used as an experimental model to study hypoxia in vitro [
17]. This model is generally used to measure neuronal viability [
27] and provides a means to measure metabolic stress and mitochondrial dysfunction as it relates to excitability of neurons [
26]. In addition, NaCN has been used to elucidate mechanisms of ischemic preconditioning [
28] and screen for potential neuroprotective compounds/drugs [
29]. Here we used NaCN combined with glucose deprivation to mimic a hypoxic ischemic environment in order to investigate the effects of DAF on neuronal injury and to explore the potential mechanisms of DAF associated with neuroprotection.
Electrophysiological changes caused by hypoxic-ischemic conditions are an early sign of neuronal injury and indicator of the degree of injury [
30]. Data recorded during cell recovery in normal medium showed that neither NaCN nor DAF changed cortical cell excitability in our model. Spontaneous neuronal electrical activity is critical for many aspects of developmental processes at all stages, such as central axon growth, navigation, and pruning of inappropriate connections [
31]. We used glutamatergic AMPA- and NMDA-mediated spontaneous plateau depolarization potentials and burst firing as an index to study neural network activities. The addition of NaCN significantly reduced neuronal spontaneous plateau potential and burst firing whereas DAF reversed these effects. This finding demonstrates that DAF promotes recovery of neuronal network dysfunction induced by hypoxia.
Dendritic spines are micron-sized protrusions of the dendritic membrane that serve as the postsynaptic component for the vast majority of central nervous system excitatory and inhibitory synapses. Spines play a crucial circuit role to synaptic matrix elements of associative neural networks and changes in dendritic spine length or shape have been shown to significantly alter the functional properties of neurons [
32]. Dendrite spines can be induced rapidly [
33] or reduced and structurally changed due to severe ischemia (within 10 min) and can be observed in the peri-infarct cortex after focal stroke in vivo [
34]. In this study, shrinkage of dendritic spines post NaCN administration was apparent as early as 10 min and significant loss of spines was evident at 60 min. Treatment with DAF resulted in a significant reduction of dendritic spine loss. These results suggest that DAF plays a role in preserving synaptic morphology during hypoxia.
It has long been believed that severe or prolonged ischemia/hypoxia leads to neuronal cell apoptosis. Consistent with previous findings [
35], when our cultured rat cortical primary neurons were exposed to chemical hypoxia, cell apoptosis was significantly increased. Indirect evidence for the participation of DAF in apoptotic events has been demonstrated in malignant tumors [
36] and neutrophils [
37]. In this study, DAF-treated cells showed an attenuated level of neuronal cell apoptosis induced by the hypoxic-ischemic insult.
We observed that cultured rat neurons constitutively possessed C3 protein which is consistent with recent findings demonstrating that C3 was present in mouse neurons [
4].
In previous studies from our laboratory, DAF was shown to inhibit formation of MAC in murine mesenteric I/R models [[
16] and unpublished data] as well as suppress production of C3a and MAC in rodent and porcine hemorrhagic shock [unpublished data]. Of particular interest is our observation that treatment with DAF significantly reduced the increase in C3 expression as well as C3a and MAC formation in hypoxic neuronal cells.
In addition to the well-described regulatory function of DAF on complement activation [
13], it has become increasingly apparent that this molecule might also act as a signal-transducing molecule. DAF has been found to associate with Src protein tyrosine kinases such as p56lck and p59fyn in human T cells [
38]. There is strong evidence demonstrating that cerebral ischemia/reperfusion induces Src activation in rat hippocampus [
39]. Furthermore, blockade or deficiency of Src activity minimized brain injury following stroke in mice [
23]. Src family kinase inhibitors PP1 and PP2 decreased brain injury induced by intracerebral hemorrhage, ischemia/reperfusion and neurosurgical procedure [
23,
40]. In the present study, we demonstrate that DAF treatment significantly decreases the activation of c-Src during ischemia-like conditions. Although the mechanism by which DAF regulates activation of Src kinase is unknown, there is increasing evidence that Src family kinases act as a point of convergence for various signaling pathway, including the pathway that signals via G-coupled receptors [
41]. Thus, it is possible that the inhibition of c-Src activity by DAF could be via a direct association of DAF and c-Src.
Neurons have been identified as the principal CNS cell that prominently expresses the C3a receptor under physiological conditions [
4]. Significant up-regulation of C3aR in murine brain after cerebral ischemia has been observed [
42]. We found that DAF dramatically diminished neuronal C3aR induced by the hypoxic-ischemic conditions. C3aR specifically binds with high-affinity to C3a and high-affinity binding sites are abundantly expressed (20,000 to 80,000 sites per cell) on cultured human astrocytes [
43]. The finding that increased C3aR was present in hypoxic cultured neurons and was associated with C3a is quite important since this can explain why soluble C3a was detected in the cell lysates. C3a-C3aR interaction was markedly reduced in the presence of DAF by limiting expression of C3aR and C3a. C3aR is a G protein-coupled receptor which initiates intracellular signaling when C3a binds to it [
44,
45]. Our findings suggest that autocrine/intracrine C3a might be involved in the regulation of neuronal functions via binding to C3aR and subsequent C3a-C3aR engagement which implies a key link to the down-stream Src and caspase signals.
Sublytic MAC may initiate cellular signal transduction pathways resulting in activation of cell survival mechanisms [
46]. However, clinical and experimental studies have implied a pivotal role for MAC in the pathogenesis of secondary neuronal cell death after TBI [
47,
48]. In our study, an increase in MAC was associated with neuronal injury suggesting lytic formation of this complex due to hypoxia is associated with cytotoxicity and subsequent cell death in this model. Taken together these observations suggest that the protective effects of DAF are related to attenuation of C3a-C3aR-Src/caspase and/or MAC-Src signaling pathways. However, this assumption is speculative and needs further investigation.
The central components of the apoptotic processes are the caspases. Cross-linking of DAF isoform with its antibody in human stomach adenocarcinoma cells elevated the expression of caspase-3 and caspase-8, and activated caspase-3 [
49]. But in hypoxic cultured neurons, we observed that application of DAF down-regulated the expression of caspase-9 and reduced caspase-3 activity. Lytic levels of MAC can trigger caspase signal pathway resulting in cell lysis or apoptosis [
50,
51] therefore it is very likely that in this model, DAF functions by downregulating caspase at least in part by blocking MAC formation. Indeed, we found that treatment with DAF diminished the colocalization/interaction of active caspase-3 and MAC caused by hypoxic conditions. This finding suggests that in addition to suppressing complement activation and Src kinase activity, DAF exerts its neuronal protective effect against hypoxia through a direct or indirect blockage of the caspase pathway.
The present study utilized cultured chemically hypoxic primary cortical neurons as a model of neuronal injury. Extrapolation of our findings to support pharmacotherapeutic innovation for the treatment of ischemic brain diseases should be weighed carefully. First, the model does not account for the role of other cellular components known to play a role in cerebral damage after ischemia and/or hypoxia. Astrocytes, oligodendrocytes and microglial cells have been reported to provide major sources of local complement activation during brain injury [
4,
8]. Second, studies on Src family kinase signaling in models of cerebral ischemia have revealed that ischemia induces an increase in tyrosine phosphorylation of n-methyl-d-aspartate (NMDA) receptors (NMDAR) by Src family kinases [
52,
53] suggesting that enhancement of Ca
2+ entry induced by the phosphorylated NMDARs or other proteins in the NMDAR complex may be important during activation of intracellular signaling cascades leading to cell death. Our study suggests that DAF interferes with complement activation, but it does not exclude the involvement of other DAF functions such as direct regulation of mitochondrial factors (e.g. apoptosis inducing factors), calcium signaling, NMDAR signaling, or actin cytoskeleton. Future studies will be necessary to determine whether DAF exerts a direct effect beyond complement inhibition on c-Src, NMDARs, transcription factors and caspases, and if so, to what extent these direct interactions contribute to the protective effects of DAF against neuronal damage during ischemia-like conditions.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
YW participated in the experimental design, performed primary cell culture, TUNEL, MTT assay, live cell image and, and Western-blot. YL participated in the experimental design, performed immunofluorescent staining experiments, data analysis, manuscript revision and formatting. SLDL revised and edited the manuscript. MS helped draft the manuscript. GCT provided critically important intellectual revision. JJDL conceived the study, participated in its design and coordination, wrote and gave final approval for manuscript submission. All authors read and approved the final manuscript.