|Home | About | Journals | Submit | Contact Us | Français|
Traumatic brain injury (TBI) activates several protein kinase signaling pathways in the hippocampus that are critical for hippocampal-dependent memory formation. In particular, extracellular signal-regulated kinase (ERK), a protein kinase activated during and necessary for hippocampal-dependent learning, is transiently activated after TBI. However, TBI patients experience hippocampal-dependent cognitive deficits that occur for several months to years after the initial injury. Although basal activation levels of ERK return to sham levels within hours after TBI, we hypothesized that activation of ERK may be impaired after TBI. Adult male Sprague-Dawley rats received either sham surgery or moderate parasagittal fluid-percussion brain injury. At 2, 8, or 12 weeks after surgery, the ipsilateral hippocampi of sham surgery and TBI animals were sectioned into transverse slices. After 2 h of recovery in oxygenated artificial cerebrospinal fluid, the hippocampal slices were stimulated with glutamate or KCl depolarization, then analyzed by western blotting for phosphorylated, activated ERK and one of its downstream effectors, the transcription factor cAMP response element-binding protein (CREB). We found that activation of ERK (p<0.05) and CREB (p<0.05) after 30 s of glutamate stimulation or KCl depolarization was decreased in hippocampal slices from animals at 2, 8, or 12 weeks after TBI as compared to sham animals. Basal levels of phosphorylated or total ERK were not significantly altered at 2, 8, or 12 weeks after TBI, although basal levels of phosphorylated CREB were decreased 12 weeks post-trauma. These results suggest that TBI results in chronic signaling deficits through the ERK-CREB pathway in the hippocampus.
There are an estimated 3.17 million people living with long-term disabilities due to traumatic brain injury (TBI) in the United States . These neurological deficits result in the inability to perform daily living activities, return to work, and participate as productive members within their communities. Longitudinal studies assessing outcome after moderate to severe TBI have consistently found that cognitive performance is impaired for months to years after the initial trauma which significantly impedes rehabilitation . Although some recovery is typically observed in the first year after injury, this either slows or reverses in the following years [23, 46]. In particular, memory problems are almost always observed in human TBI patients, due either to direct effects on memory encoding or secondary effects on concentration and attention [1, 30]. This is reflected in experimental models of TBI, which have reported enduring impairments in learning and memory as well [10, 24, 35].
Functional imaging studies in humans have found that underlying the neurological deficits in cognition is atrophy of the hippocampus and white matter tract damage [7, 41, 47]. Furthermore, there is neuronal loss throughout the CA1, CA3 and CA4 subregions of the hippocampus . Besides these histopathological changes, whether the remaining hippocampal neurons exhibit persistent deficits in cell signaling pathways required for learning and memory after TBI remains relatively unexplored.
Using the parasagittal fluid-percussion brain injury (FPI) as a clinically relevant model of TBI, we and others have found that TBI activates a series of biochemical cascades in the hippocampus; several of these signaling pathways are critical for the formation of hippocampal-dependent memories. The protein kinases extracellular signal-regulated kinase (ERK), calcium/calmodulin-dependent protein kinase II (CaMKII), protein kinase C (PKC), and calcium/calmodulin-dependent protein kinase IV (CaMKIV) are all activated after TBI [2, 15, 19, 21, 25, 34, 36, 50]. However, these changes are rapid and transient, occurring on the timescale of minutes to hours and reflect the glutamate excitotoxicity and widespread potassium depolarization that initially occurs as a consequence of the trauma . Over the course of hours to days, activation of these signaling molecules returns to noninjured levels. Although the acute activation of protein kinases resolves between hours to days after TBI, we hypothesized that activation of these signaling pathways is chronically impaired after TBI.
All procedures with animals were in accordance with the NIH Guide for the Care and Use of Laboratory Animals and approved by the University of Miami Institutional Animal Care and Use Committee. Six experimental groups were used in the following experiments: sham surgery animals (2 week survival n=4, 8 week survival n=4, 12 week survival n=4) and TBI animals (2 week survival n=6, 8 week survival n=6, 12 week survival n=6). Male Sprague Dawley rats (270–320 gm) were anesthetized with 3% halothane, 70% N2O, and 30% O2 and received a 4.8 mm craniotomy (3.8 mm posterior to bregma, 2.5 mm lateral to the midline) to secure a plastic tube (3.5 mm inside diameter) over the right parietal cortex with cyanoacrylate and dental cement. Twenty-four h after the craniotomy, the animals were re-anesthetized, intubated and mechanically ventilated with 0.5–1% halothane, 70% N2O, and 30% O2. Pancuronium bromide (0.5 mg/kg, intravenously) was administered to immobilize the rats. Arterial blood pressure, blood gases, and blood pH were monitored for 30 min prior to and up to 30 min after TBI to maintain physiological ranges (blood pH: 7.35–7.45, pCO2: 35–40 mm Hg, pO2: 105–140 mm Hg). Brain and body temperature were monitored with thermistors placed in the left temporalis muscle and rectum and maintained at 36.5–37.0°C with self-adjusting warming lamps. The animals received a moderate (2.0 ± 0.2 atmospheres) fluid-percussion pulse or sham treatment.
At 2, 8, or 12 weeks of recovery after surgery, animals were anesthetized with 3% halothane, 70% N2O, and 30% O2 for 5 min, and then decapitated. The ipsilateral hippocampus was rapidly dissected and transversely sectioned into slices 400 μm thick at 4°C using a vibratome in artificial cerebrospinal fluid (aCSF, in mM: 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 25 NaHCO3, 10 D-glucose, 2 CaCl2, and 1 MgCl2, saturated with 95% O2/5% CO2). An average of 8–12 hippocampal slices were obtained from each sham and TBI animal. After 1 h of recovery at room temperature, slices were equilibrated in a submersion chamber at 31.5°C for 1 h. Hippocampal slices (n=2–4 for each treatment condition from each animal) were pharmacologically stimulated with glutamate (200 μM) or high KCl depolarization (90 mM) for 30 s in isotonic aCSF. After stimulation, the slices were transferred to a nylon screen and frozen on liquid nitrogen. Two hippocampal slices from each treatment condition (control, high KCl depolarization, or glutamate) were pooled and sonicated in 200 μl of fractionation homogenization buffer (15 mM Tris pH 7.6, 0.25 M sucrose, 1 mM MgCl2, 1 mM EGTA, 1 mM DTT, 1.25 μg/ml pepstatin A, 10 μg/ml leupeptin, 25 μg/ml aprotinin, 0.5 mM PMSF, 0.1 mM Na3VO4, 50 mM NaF, 2 mM Na4P2O7, and 1X phosphatase inhibitor cocktail set II, Calbiochem, San Diego, CA, USA). Samples were electrophoresed and western blotted for anti-phospho-ERK2 Thr202/Tyr204 (1:3000, Cell Signaling Technology, Danvers, MA, USA), anti-total ERK2 (1:5000, Cell Signaling Technology), anti-phospho-CREB Ser133 (1:1000, Cell Signaling Technology), anti-total CREB (1:1000, Cell Signaling Technology), and β-actin (1:10,000, Sigma-Aldrich, St. Louis, MO, USA). Epitopes were visualized with HRP-conjugated secondary antibodies (1:2000, Cell Signaling Technology) using enhanced chemiluminescence and developed on film (Phenix x-ray film BX; Phenix Research Products, Hayward, CA, USA). Films were developed to be in a linear range and densitized using ImageJ 1.38x (National Institutes of Health, USA). To analyze changes in phospho-ERK and phospho-CREB after glutamate or high KCl stimulation, levels of phospho-proteins from stimulated slices were normalized to levels phospho-proteins from non-stimulated slices within each animal. Thus, non-stimulated phosphorylation levels from both sham and TBI hippocampal slices were averaged to 1.0 to directly compare the changes in phosphorylation after high KCl depolarization or glutamate stimulation. To analyze changes in basal levels of phospho-ERK and phospho-CREB, levels of phospho-proteins from all TBI animals were normalized to levels of phospho-proteins from all sham surgery animals within each survival time point.
All of the data presented are mean ± SEM. Statistical analysis was performed with GraphPad Prism 5.02 (GraphPad Software, Inc., La Jolla, CA, USA). The differences among multiple groups were analyzed with a two-way analysis of variance (ANOVA) with significance set at p<0.05. Post-hoc comparisons were made using Tukey t-tests.
To investigate changes in protein kinase signaling after TBI, we dissected the ipsilateral, injured hippocampus from sham and TBI animals, and generated hippocampal slices. An average of 8–12 hippocampal slices were generated from each animal. After 2 h of recovery in oxygenated aCSF, we measured levels of phosphorylated, activated ERK in hippocampal slices after brief high KCl depolarization (90 mM) or glutamate stimulation (200 μM) for 30 s. These stimulation paradigms trigger a transient depolarization and brief calcium influx into hippocampal neurons mediated through voltage-gated calcium channels and glutamate receptors [5, 12, 13]. We found that there was a significant increase in ERK activation with KCl treatment or glutamate treatment in hippocampal slices from sham surgery animals, but not from TBI animals at 2 weeks after surgery (Fig. 1) [injury: F(1,46) = 33.87, p<0.0001, slice treatment F(2,46) = 18.51, p<0.0001]. This was seen as well at 8 weeks after surgery [injury: F(1,45) = 40.63, p<0.0001, slice treatment F(2,45) = 14.40, p<0.0001] and 12 weeks post-injury, although there was some variability in the response to stimulation in TBI hippocampal slices [injury: F(1,39) = 17.15, p=0.0002, slice treatment F(2,39) = 9.27, p=0.0005].
A downstream effector of ERK during hippocampal-dependent learning is the transcription factor CREB [42, 44, 48]. After stimulation with high KCl depolarization or glutamate (200 μM) for 30 s, hippocampal slices were analyzed by western blotting for changes in CREB phosphorylation (Fig. 2). We found that activation of CREB was reduced in hippocampal slices from TBI animals as compared to slices from sham animals at 2 weeks [injury: F(1,46) = 21.34, p<0.0001, slice treatment F(2,46) = 9.30, p=0.0004], 8 weeks [injury: F(1,46) = 34.66, p<0.0001, slice treatment F(2,46) = 15.90, p<0.0001], and 12 weeks post-injury [injury: F(1,38) = 9.67, p=0.0035, slice treatment F(2,38) = 13.47, p<0.0001].
To evaluate whether basal levels of ERK or CREB could account for the deficits in activation, we directly compared basal levels of phosphorylated and total ERK and CREB from non-stimulated hippocampal slices of sham surgery or TBI animals. To control for changes in total protein levels, we normalized the changes in phospho- and total ERK and CREB levels to β-actin levels within each sample. Basal levels of ERK activation or total levels of ERK in hippocampal slices were not significantly different between sham and TBI animals at 2 weeks or 8 weeks post-TBI (Fig. 3), although a nonsignificant trend towards a decrease in ERK phosphorylation was observed at 12 weeks post-TBI [injury: F(1,33) = 0.41, p=0.5256, recovery time F(2,33) = 1.84, p=0.1740]. Basal levels of CREB activation or total CREB levels were not significantly decreased at 2 or 8 weeks post-injury (Fig. 4). At 12 weeks after injury, there was a significant decrease in basal phospho-CREB levels in hippocampal slices from TBI animals as compared to sham animals [injury: F(1,31) = 11.48, p=0.0019, recovery time F(2,31) = 4.16, p=0.0250, interaction: : F(2,31) = 4.18, p=0.0246].
After TBI, there are both acute and chronic behavioral deficits [23, 30, 46]. With the fluid-percussion brain injury model, we can reliably observe deficits in hippocampal-dependent learning that persist from days to months after trauma [10, 11, 22, 29, 35]. There are a number of mechanisms that may contribute to hippocampal-dependent learning deficits. Accordingly, there is hippocampal atrophy, neuronal and synaptic loss in select regions of the hippocampus, atrophy of white matter tracts, as well as changes in inputs from other brain regions into the hippocampus [8, 9, 17, 20, 38, 49]. Our results suggest that another mechanism that may underlie chronic deficits in learning after TBI is the inability to activate the ERK-CREB signaling pathway in remaining hippocampal neurons.
Although we normalized the changes in phospho-ERK and phospho-CREB to total protein levels, and total ERK, total CREB or β-actin levels were not significantly different between sham and TBI injured animals, the number of synapses and neurons activated in a hippocampal slice from sham animals is likely to be much greater than from TBI animals [8, 9, 20, 39, 49]. Thus, the deficits in ERK and CREB activation may be due to neuronal and synaptic loss which results in atrophy of the hippocampus in the weeks following injury. Whether deficits in this signaling pathway are seen in mild fluid-percussion brain injury where there is no significant cell loss or atrophy of the hippocampus remains to be determined [21, 29]. Evaluation of injury severity may differentiate the contribution of neuronal loss to the misregulation of the ERK-CREB signaling pathway after TBI and is an important area of future investigation. Furthermore, there is select neuronal loss in the CA3 region of the hippocampus and dentate hilar region as compared to other subregions of the hippocampus with the parasagittal fluid-percussion brain injury model [9, 20, 49]. Whether there are greater deficits in ERK and CREB activation in these areas or whether this occurs throughout all subregions of the hippocampus remains to be explored .
TBI results in significant impairments in the ability to maintain calcium homeostasis . Accordingly, although glutamate elicits similar levels of peak calcium entry into CA3 hippocampal neurons at 4 weeks after TBI, the restoration of basal calcium levels in CA3 neurons from TBI animals is impaired [16, 45]. This suggests that the molecular mechanisms that maintain homeostatic calcium levels are chronically altered after TBI. However, since the peak calcium influx was comparable in cultured neurons from sham and TBI animals and we assessed ERK and CREB activation during the peak calcium influx, it is unlikely that the deficits we observed are due to changes in total calcium levels during the glutamate pulse or potassium depolarization.
Given that we observed deficits in ERK activation with glutamate or potassium depolarization, the synaptic mechanisms that may underlie the deficits in ERK activation may involve changes in glutamate receptors and/or voltage-gated calcium channels. There are transient, biphasic changes in NMDA receptor levels after TBI, but by 14 days post-injury, these changes have returned to sham levels [6, 27, 32]. Levels of the GluR1 subunit of the AMPA receptor are increased 3 days post-TBI ; however whether these changes are persistent or if there are any differences in voltage-gated calcium channel expression at chronic time points after brain trauma remains unknown. Underlying these biochemical changes are deficits in hippocampal synaptic plasticity. Previous studies have found that there are significant impairments in long-term potentiation (LTP) after TBI, a physiological correlate of learning and memory formation [14, 38, 40, 43]. Inhibition of ERK activation blocks maintenance of hippocampal LTP [4, 18], which mimics the deficits in hippocampal LTP seen after TBI. Thus, one potential mechanism for the deficit in hippocampal LTP after TBI may be the inability of the glutamatergic synapse to activate ERK and trigger CREB-mediated gene expression.
ERK is activated by a canonical pathway that involves B-Raf activation of mitogen-activated protein kinase kinase (MEK) which then phosphorylates and activates ERK. Another possible mechanism for the impairments in ERK activation is the inability to stimulate one or several upstream activators of ERK. However, a causal link between the deficit in ERK and CREB activation after TBI remains to be established. CREB is phosphorylated by a number of protein kinases downstream of ERK, such as p90 ribosomal S6 kinase, mitogen- and stress-activated kinase 1/2, but is also directly phosphorylated by CaMKIV and protein kinase A (PKA) . We have previously demonstrated that cAMP levels and PKA activation decrease from 15 min to 24 h after TBI in the injured hippocampus . Although cAMP levels return to non-injured levels at 48 h after TBI, PKA activation is still depressed at 48 h post-TBI. Furthermore, whether there is a decrease in cAMP levels or PKA activation in the weeks after brain injury is unknown. A decrease in either cAMP levels or PKA activation could underlie the deficits in ERK and/or CREB activation. Thus, the deficits in CREB activation may have occurred through chronic alterations in one or several protein kinase pathways. An alternative mechanism that could account for the deficits in ERK and CREB activation is increased phosphatase activity. Although persistent alterations in phosphatases that dephosphorylate ERK have not yet been examined after TBI, calcineurin activity is increased 2–3 weeks after TBI .
Understanding the molecular mechanisms that underlie enduring deficits in learning and memory after TBI may open new therapeutic avenues to pursue to improve cognition after TBI. In the current study, we have identified deficits in the ability of the hippocampus to activate signaling through ERK and CREB after brain trauma. Development of pharmacological strategies to stimulate signaling through ERK and CREB may thus be a therapeutically feasible strategy in the rehabilitation repertoire for TBI patients.
This work was supported by USAMRMC PR054538 and NIH grant NS056072.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.