|Home | About | Journals | Submit | Contact Us | Français|
Calcineurin (CaN) is a calcium/calmodulin-dependent phosphatase directly activated by calcium as a result of neuronal activation that is important for neuronal function. CaN subunit isoforms are implicated in long-term potentiation (LTP), long-term depression (LTD), and structural plasticity. CaN inhibitors are also beneficial to cognitive outcomes in animal models of traumatic brain injury (TBI). There are known changes in the CaN A (CnA) subunit following fluid percussion injury (FPI). The CnA subunit has two isoforms: CnAα and CnAβ. The effect of moderate controlled cortical impact (CCI) on distribution of CnA isoforms was examined at 2h and 2 weeks post-injury. CnA distribution was assayed by immunohistochemistry and graded for non-parametric analysis. Acutely CnA isoforms showed reduced immunoreactivity in stratum radiatum processes of the ipsilateral CA1 and CA1–2. There was also a significant alteration in the immunoreactivity of both CnA isoforms in the ipsilateral dentate gyrus, predominantly within the hidden blade. Alterations in CnA isoform regional distribution within the CA1, CA1–2, and dentate gyrus may have significant implications for persistent hippocampal dysfunction following TBI, including dysfunction in hippocampal plasticity. Understanding alterations in CnA isoform distribution may help improve the targeting of current therapeutic interventions and/or the development of new treatments for TBI.
The neurological mechanisms that contribute to cognitive and behavioral deficits following traumatic brain injury (TBI) are not well understood. Protein phosphorylation is a dynamic, rapidly reversible post-translational modification, known to be involved in neuronal activation, plastic remodeling, and memory.
One of the most abundant and the only calcium-activated phosphatase in the brain is protein phosphatase 2B, alternatively know as calcineurin (CaN) (Klee et al., 1979). CaN is a calcium/calmodulin-dependent phosphatase that is highly sensitive to, and preferentially activated by, minor changes in intracellular calcium (Ca2+) (Rusnak and Mertz, 2000). CaN dephosphorylates several key cytoskeletal and synaptic vesicle proteins (Rusnak and Mertz, 2000), modulates the activity of the transcription factors nuclear factor of activated T-cells and cAMP response element-binding protein (Yang and Klee, 2000), and regulates neuronal excitability through modulation of γ-aminobutyric acid (Huang and Dillon, 1998) and N-methyl-D-aspartate acid (NMDA) receptors (Tong et al., 1995).
The phosphorylation state of AMPA and NMDA receptors is regulated by the interaction of protein kinase A and CaN (Banke et al., 2000; Beattie et al., 2000), which directly affects long-term potentiation (LTP) (Lee et al., 2003) and long-term depression (LTD) (Lee et al., 2002), cellular functions linked to memory and learning (Li et al., 2003; Lemon and Manahan-Vaughan, 2006). CaN, through its interaction with kinase-anchoring proteins, has also been shown to regulate structural proteins important for synaptic remodeling (Dell'Acqua et al., 2006).
Pathologically, excessive CaN has been linked to mitochondrial dysfunction and apoptosis (Asai et al., 1999). Increases in basal and maximal CaN activity (Kurz et al., 2005b), and alterations in CaN subunit A subcellular distribution (Kurz et al., 2005a), have been reported in the hippocampus following fluid percussion (FP) TBI, and they can persist for 2–3 weeks following injury. Increases in CaN activity following TBI may be due to significant increases in intracellular Ca2+ that occur early in brain injury (Shapira et al., 1989; Fineman et al., 1993). CaN activity following TBI is generally considered to be pathologically altered and has been associated with increases in cellular death and dysfunction in both ischemia and TBI (Morioka et al., 1999). Inhibition of CaN activity is neuroprotective in ischemia and TBI models (Butcher et al., 1997; Friberg et al., 1998; Okonkwo et al., 1999, 2003; Mbye et al., 2009).
CaN is composed of a 57–61-kDa catalytic subunit (CnA), and a 19-kDa regulatory subunit (CnB) (Klee et al., 1979; Rusnak and Mertz, 2000, Groth et al., 2003). Both the CnA and CnB subunits consist of several different isoforms (CnAα, CnAβ, and CnAγ, and CnB1 and CnB2, respectively), which may offer unique substrate specificity, localization, and/or recognition properties to the phosphatase. The overexpression of CnA (Asai et al., 1999) has been shown to cause apoptosis through CaN-dependent dephosphorylation of BAD (Wang et al., 1999), and also cause mitochondrial dysfunction, elevated superoxide levels, and increased production of reactive oxygen species (Manalan and Klee, 1983; Morioka et al., 1999; Wu et al., 2004). The levels of the CnA have also been found to change in synaptic membrane fractions for up 2 weeks following fluid percussion injury in the rat (Kurz et al., 2005a). Divergent changes were found in CnAα and CnAβ after ischemia in the gerbil (Hashimoto et al., 1998) and in CnAα, CnAβ, and CnAγ in schizophrenia (Liu et al., 2007). CnA subunits are particularly important for the initiation and maintenance of LTP (Wang and Kelly, 1996; Kayyali et al., 1997; Winder and Sweatt, 2001), and play an active role in structural plasticity of cortical circuits (Victor et al., 1995; Yakel, 1997). Following TBI there is a loss of LTP that persists for up to 2 weeks post-injury (Reeves et al., 1995; D'Ambrosio et al., 1998; Sick et al., 1998; Sanders et al., 2000).
Although CnA isoforms have been located in brain tissue, little is known about the relative expression of the CnA isoforms in brain and how they are altered by trauma. The aim of this study was to determine how the CnA isoforms are distributed in the rat hippocampus, and to elucidate how CnA isoforms are affected acutely and chronically following TBI.
Adult male Sprague-Dawley rats (n=30) were used in the study. The rats were purchased from Hilltop Laboratories (Scottsdale, PA) and housed in pairs under a 12-h:12-h light:dark cycle. Rats were given food and water ad libitum throughout the study. All experiments were carried out in accordance with the University of Pittsburgh's guidelines for the Care and Use of Laboratory Animals. All experiments were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh.
On the day of surgery anesthesia was initiated with 4% isofluorane (IsoFlo; Abbott Laboratories, North Chicago, IL) and 2:1N2O:O2. Rats were intubated and maintained on 1.5–2% isofluorane during the surgical procedure. Following intubation, the rat was placed on a thermal blanket to regulate body temperature (37°C) and the animal's head was placed in a stereotaxic frame. An incision was made down the midline of the skull and the soft tissues and periosteum deflected. A craniotomy was then performed over the right parietal bone to expose the dura. Controlled cortical injury (CCI; Pittsburgh Precision Instruments, Inc., Pittsburgh, PA) at a depth of 2.4mm at 4m/sec was carried out according to the Dixon method (1991). A total of 18 rats were injured, and the remaining 18 rats were shams. The righting reflex (Dixon et al., 1991) was monitored in the immediate post-surgical period to assess acute recovery.
Following a 2-h (6 sham animals and 6 TBI animals), or 2-week (6 sham animals and 6 TBI animals) recovery period, the animals were given an overdose of sodium pentobarbital (100mg/kg IP), and perfused intra-aortically with 0.1M heparinized PBS in 4% PFA/0.1M PBS. The brains were dissected, submerged in increasing concentrations of sucrose, and stored at −80°C. The brains were then sectioned at 35μm in a cryostat, and free-floating sections were collected in tissue plate wells containing 0.1M TBS (pH 7.5).
All immunohistochemical procedures and incubations were carried out with agitation with the exception of the chromogen step. All treatment groups were stained together in each immunohistochemical session. The sections were matched by region, rinsed 3×5min in washing buffer (0.1% Triton-X in 0.1M TBS) and blocked in a mixture of 10% normal donkey serum in washing buffer for 2h at room temperature. The sections were then incubated overnight at 4°C in (one only) primary goat antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) specific to the CnAα (1:150), CnAβ (1:150), CnAγ (1:100) subunit isoforms diluted in washing buffer/5% normal donkey serum. Following incubation, the sections were then washed 3×8min in washing buffer and endogenous peroxidase activity was quenched with 0.3% H2O2 in methanol for 10min. Following 5×5min washing in TBS, the sections were incubated for 1h at room temperature in horseradish peroxidase-conjugated donkey anti-goat secondary antibody (Jackson ImmunoResearch, West Grove, PA) at 1:200 in 0.1M TBS. Immunoreactivity (IR) was then visualized using 0.01% 3′-3′-diaminobenzidine (DAB) after an extensive wash. The DAB reaction was terminated with dH2O and sections were rinsed in 0.1M TBS, mounted onto slides, air dried, and cover-slipped for light microscope analysis. All sections within the reaction were exposed to each of the reagents for the same time period. Control sections pre-adsorbed with their homologous peptides were negative for IR.
At 2h or 2 weeks (3 sham and 3 TBI per time point) animals were deeply anesthetized with pentobarbital (Nembutal, 80–100mg/kg; Abbott Laboratories). The animals were decapitated and the brains quickly removed and chilled on ice. Both the right and left hippocampi were exposed by careful dissection and the dentate gyrus (DG) from each side was excised. Following the removal of the DG the remainder of the hippocampal formation was cut, removing the CA1 region from the CA1–2/CA3 regions. The DG, CA1 region, and CA1–2/CA3 regions from three animals were combined and collected into separate tubes for each region. Tissues from the hippocampus ipsilateral to injury and the hippocampus contralateral to injury were collected separately. Tissue was immediately placed in liquid nitrogen, and then into a −80°C freezer until processing. Tissue was homogenized in lysis buffer (suspension buffer), which contains 0.1M NaCl, 0.01M Tris-Cl (pH 7.6), 0.001M EDTA (pH 8.0), 1μg/mL aprotinin, 100μg/mL phenylmethylsulfonyl fluoride, and protein concentrations were determined using a BCA protein assay kit (Pierce, Rockford, IL). Samples containing 20μg of protein were subjected to SDS-polyacrylamide gel electrophoresis through a 10% acrylamide gel, and then transferred to nitrocellulose membranes and immunolabeled with antiserum, followed by donkey anti-goat immunoglobulin G conjugated to peroxidase (1:5000; Pierce). Proteins were visualized with a chemiluminescence detection system (SuperSignal; Pierce). To assure equal loading, the membrane was re-striped and re-blotted with rabbit anti-actin antibody (1:15,000; Sigma, St. Louis, MO).
Two hours corresponds to the acute alterations seen after TBI, including rapid increases in Ca2+ that occur immediately post-injury and persist (Hovda et al., 1990; Shapira et al., 1989). Given the role of Ca2+ in the control of CaN activity, an acute examination of isoform alterations at 2h post-injury was chosen to examine the effect of acute Ca2+ increases. Two weeks corresponds to the time point used to assess cognitive outcomes following CCI in rats (Yan et al., 2000; Kline et al., 2002).
A 6-point scoring scale (0=no IR to 6=heavy IR) was utilized to assess the level of CnA isoform expression in DAB-stained sections from separate subregions (CA1, CA1–2, CA3, dentate gyrus hilum, dentate gyrus hidden blade, and dentate gyrus exposed blade) of the ipsilateral and contralateral hippocampi. To assess dendritic staining in the stratum radiatum of the ipsilateral and contralateral CA1 and CA1–2 a 3-point scoring scale was employed (1=few to 3=numerous). An independent observer blinded to time point and isoform stain scored each tissue section utilizing both scoring paradigms. Following scoring the data were organized by CnA isoform stain, time point (2-h or 2 week), and ipsilateral or contralateral. Scores were ranked (blinded to sham or TBI) from 1–12 (ipsilateral separate from contralateral) for each subregion and dendrite score. Following ranking, sham and TBI designations were revealed. Outcomes of sham versus injury were compared using the Wilcoxon-Mann-Whitney two-sample rank-sum test with a p value for significance set at ≤0.05.
CnA isoform distribution within the rat hippocampus was determined utilizing immunohistochemical staining in sham rats. Both isoforms CnAα (sham groups in Figs. 1 and and3)3) and CnAβ (sham groups in Figs. 4 and and6)6) showed similar distribution patterns, with higher expression in the CA1 neuropil and cell bodies than CA1–2 regions, and noticeable columnar patterns of pyramidal neuron dendritic staining within the stratum radiatum of both the CA1 and CA1–2. Within the CA3 the CnAα isoform (sham groups in Figs. 1 and and3)3) appears to be predominantly in the stratum lucidum, with little expression within cell bodies of the stratum pyramidale layer. However the CnAβ isoform (sham groups in Figs. 4 and and6)6) is not as layer-specific within the CA3 region, and appears to have more widespread expression throughout the layers of CA3. In general the CnAβ isoform (sham groups in Figs. 4 and and6)6) shows relatively lower expression within the CA1, CA1–2, and CA3 regions of the hippocampus compared to the CnAα isoform (sham groups in Figs. 1 and and3).3). However, both isoforms (sham groups in Figs. 1–6) show marked staining within the exposed blade of the DG in both the cell bodies of the stratum granulosum and neuropil of the stratum moleculare layers. There was relatively little expression within the DG hidden blade (HB) or hilum of either the CnAα (sham groups in Figs. 1 and and3)3) or CnAβ isoforms (sham groups in Figs. 4 and and6).6). The CnAγ isoform showed no distribution within the rat hippocampus (not pictured).
There were clear changes in CnA isoform distribution in the rat hippocampus at 2h post-TBI, with persistent alterations lasting until 2 weeks post-TBI, as determined by sham versus injury comparison using the Wilcoxon-Mann-Whitney two-sample rank-sum test with significance set at p≤0.05 (Table 1).
CnAα staining in the injured ipsilateral CA1 region was in cell bodies within the stratum pyramidale (SP) layer, and in the stratum oriens (SO) layer, without localization to cell bodies and no clear dendritic distribution (Fig. 1). There was a loss of the columnar distribution of CnAα staining within dendrites of the stratum radiatum (SR) layer compared to sham animals (Fig. 2). Within the contralateral CA1 of injured animals there was a noticeable difference in cellular distribution of the CnAα isoform (not pictured), with significant staining of dendrites and cell bodies throughout the SR and SP layers that is similar to sham animals. There appears to be a general decrease in contralateral expression of CnAα within the CA1 and an increase in contralateral expression of CnAα within the exposed blade of the DG in injured versus sham animals that did not reach statistical significance (Fig. 1).
The majority of CnAα staining in ipsilateral CA1–2 and CA3 regions of injured animals were in the surrounding neuropil of the SO and SR, with a few cell bodies within the SP showing expression that was not seen in sham animals. There was also decreased dendritic staining in the SR of the CA1–2 region compared to sham animals. Increases in staining within the HB and hilum of the DG were predominantly in the neuropil, with some localization to cell bodies (the stratum granulosum [SG] in the HB). Decreases in CnAα staining within the exposed blade of the DG was localized to cell bodies within the SG layer and neuropil of the stratum moleculare (SM) layer (Fig. 2). The loss of CnAα isoform expression within the exposed blade of the DG was predominantly lateral to the midline, with a relative sparing of expression in the genu of the DG.
At 2 weeks post-injury, staining within the ipsilateral CA1 appeared to be localized to the SP layer within the cell somas (Fig. 3). In many sections there was a loss of CnAα dendritic expression within the ipsilateral CA1, consistent with what occurred at 2h post injury; however, the variability in dendritic expression among animals did not reach statistical significance in SR dendrite staining of the CA1 in injured hippocampi compared to sham animals. CnAα expression in injured CA1–2 and CA3 regions of the ipsilateral hippocampus was in the neuropil of the SO and SR. Similarly to the ipsilateral CA1–2 region at 2h, at 2 weeks very little CnAα staining of dendrites could be appreciated within the SR. There was also a loss of staining within the stratum lucidum (SL) layer of the ipsilateral CA3 region at 2 weeks compared to sham animals that was not seen at 2h.
There was higher expression in the hilum and HB of the injured ipsilateral DG in both cell somas and surrounding neuropil compared to sham animals; however, it did not reach statistical significance. In the exposed blade of the ipsilateral DG CnAα demonstrated reduced expression, predominantly within cell somas of the SG, with some decreases in the surrounding neuropil compared to sham animals. Similarly to 2h post-injury, there remained relatively high expression of CnAα within the genu of the ipsilateral DG.
Acute changes in staining of the CnAβ subunit were similar to alterations seen with CnAα subunit staining. CnAβ staining was in the neuropil of the SO and SR of both the CA1 and CA1–2 regions, with noticeably less dendritic staining in the SR of both regions (Fig. 4). Unlike CnAα there did not appear to be as much distribution of CnAβ to the cell somas of the SP layer in either CA1 or CA1–2. CnAβ distribution within the CA3 region was distinct from that seen with CnAα, as there was more CnAβ staining within the cell somas of the SP layer of CA3 that was not as apparent following injury; however, it was not a significant difference. There were non-significant decreases in CnAβ staining within the SL of CA3, and increases within the SO layer. There was less CnAβ staining of both the cell somas of the SG and the neuropil of the SM layer in the exposed blade of the injured ipsilateral DG compared to sham animals. Similarly to CnAα, there was also a non-significant decrease in CnAβ expression within both the CA1 and CA1–2 regions that only appeared to occur at the acute time point.
In sham animals at 2 weeks post-injury, the staining of the CnAβ subunit in the CA1 region of both hippocampi appeared to be less than that appreciated at 2h post-injury in sham animals. There was less CnAβ staining within dendrites of the SR of both the CA1 and CA1–2 regions, although neither reached statistical significance owing to the variability of dendrite expression in injured CA1 and CA1–2 at 2 weeks post-injury (Fig. 6). At 2 weeks changes within the CA3 region of the CnAβ subunit mirrored those seen at 2h, with apparently less staining within the cell somas of the SP layer and increases within the neuropil of the SO and SR layers, which did not reach statistical significance. Similarly to 2h post-injury, there was less CnAβ expression within the exposed blade of the injured ipsilateral DG compared to sham animals, predominantly in the SG layer.
Western blot analysis of ipsilateral and contralateral DG, CA1, and CA1–2/CA3 regions indicated no difference between sham and injured animals in the protein concentrations of the CnAα isoform relevant to each region. This suggests that the alterations in CnAα isoform staining seen in immunohistochemistry were due predominantly to a redistribution in the regions of interest, and not due to altered protein synthesis (Fig. 7). However, there was an apparent decrease in protein expression of CnAβ in the ipsilateral DG (Fig. 7), which was consistent with the immunohistochemistry. There were no other apparent differences in CnAβ protein expression in other regions.
Following CCI there are regionally-specific alterations in CnA subunit isoform distribution. Acutely there is decreased IR in both CnAα and CnAβ within the exposed blade of the DG, and a decrease in dendritic IR within both the CA1 and CA1–2 regions of the ipsilateral hippocampus in CCI versus sham animals. Alterations in IR within the DG persist chronically, as do changes in dendritic IR, although not consistently.
In the acute phase after TBI, elevations in Ca2+ (Shapira et al., 1989) can lead to abnormal neuronal activation, leading to excitoxicity, increased superoxide levels, oxidative stress, and cell death (Hovda et al., 1992). Although CaN appears be a key mediator in many of these processes, there needs to be a greater understanding of how CaN is specifically modulated after injury, particularly with regard to the individual CaN subunits and isoforms. The present study showed that CnA isoforms are differentially modulated in the hippocampus in a regionally-specific manner after TBI. Indeed, both up- and downregulation of the same CnA isoform can occur in regionally distinct areas of the hippocampus within the same post-injury time period. This suggests that individual CaN isoforms may have different roles within select hippocampal areas in the post-injury period. Whether these changes are a short-term response to injury, the initiation of long-term neuronal destabilization mechanisms, or a counter-regulatory mechanism awaits further studies conducted in a temporal manner using specific CaN inhibitors.
The present study illustrates that changes in CaN appear as early as 2h post-TBI and persist until at least 2 weeks post-injury. Changes in hippocampal IR as assessed utilizing immunohistochemical staining of CnA isoforms were noticeable in all but the CnAγ isoform, which had little to no expression within the hippocampus. Western blot data suggest that the visualized alterations in IR were not associated with changes in protein concentrations within gross hippocampal regions, but rather in cellular distribution or perhaps in divergent expression within subsets of different cell types throughout the hippocampus (Fig. 7). Blinded scoring of immunohistochemical staining suggests that there may be some expression changes within regions of the different areas of the hippocampus, including persistent decreases in the EB of the dentate gyrus of both CnA isoforms (Table 1). The discrepancy between Western blot data and immunohistochemistry is most likely due to the fact that Western blots relied upon excised tissue sections that could not separate subsections of hippocampal regions, and instead only analyzed gross dissections of the dentate gyrus, CA1, and CA2–CA3 regions.
The expression of the CnA subunit has been linked to mitochondrial dysfunction and cell death signaling through BAD (Asai et al., 1999). CnA isoform expression, specifically CnAα, is known to be important in both oxidative stress and mitochondrial dysfunction, which may play a role in both acute and chronic dysfunction within the hippocampus following TBI (Uchino et al., 2008). The level of cellular death was not assessed in this injury paradigm. However, given the regionally specific and variable alteration of CnA cellular distribution within the CA1, CA1–2, CA3, and the hilum and HB of the DG acutely and chronically, there may be broader implications than simply cell death signaling (Figs. 1–6).
Another possible explanation for changes in CnA isoform subunit distribution is related to the structural function of hippocampal cell signaling (Buzsaki, 1996; Muller et al., 1996). The entorhinal cortex (EC) sends projections through two separate pathways into the hippocampus. The perforant pathway synapses within the DG and CA3 regions, which then send projections into the CA1 region, while another separate pathway from the EC directly synapses within the CA1 region. The CA1 then sends its axonal outputs back into the deep layers of the EC.
Both the DG and CA1 dendritic field receive input from the EC through two separate pathways. Alterations in isoform IR distribution in these regions may be due to excitatory glutamate release from cortical structures occurring post-injury. Isoform alterations in contralateral isoform expression appreciated acutely in both the CA1 and DG (Figs. 1 and and4),4), that are no longer present at 2 weeks post-injury (Figs. 3 and and6),6), suggest a potential cortical component. Given that there is extensive crossover signaling between the right and left EC and hippocampus, excitotoxic glutamate release could explain isoform alterations in the areas that receive cortical input, either ipsilateral or contralateral to injury (the DG and CA1 regions) (Buzsaki, 1996). Sustained low levels of Ca2+ caused by increased cellular excitability due to increased glutamate release in the post-injury period may activate CnA and be responsible for the persistent changes in isoform IR distribution in the CA1, CA1–2, and DG regions. Alterations in Ca2+ concentration and flow, both acutely and chronically, have been appreciated within the hippocampus following TBI (Deshpande et al., 2008; Sun et al., 2008).
At 2 weeks, alterations in CnA isoform distribution within the SR of CA1 and CA1–2 did not reach significance (Figs. 3 and and6).6). This may indicate that after a period of recovery there is a reversal of initial alterations in isoform distribution and a possible recovery of CaN subunit function chronically. This may be partially responsible for the recovery seen post-TBI. This was not seen in the dentate, however, and acute alterations in isoform distribution persisted into the chronic phase. What consequences regionally-specific changes have on global hippocampal function is unclear; however, given the extensive cortical connections throughout the dentate it is possible that persistent ipsilateral dentate changes in the localization of CnA are related to changes in synaptic inputs from cortical neurons.
Changes in CnA distribution and expression profiles throughout the hippocampus have important implications for cellular function independent of cellular death pathways. There is a consistent decrease in dendrite IR of both isoforms within the ipsilateral CA1 and CA1–2 regions at 2h, which partially reverses by 2 weeks (Figs. 1–6). This is in contrast to prior results demonstrating persistent increases in CnAα isoform staining within the apical dendrites of CA1 neurons (Kurz et al., 2005a). However, it must be noted that in the previous study the authors used an FP model, and thus there may be different alterations in CaN distribution with CCI. It is also important to recognize that while there is reduced dendritic staining in the deep layers of the CA1 and CA1–2 SR, there remains significant CnAα IR in areas closer to the cell soma, which is consistent with the results presented by Kurz and associates (2005a).
Several studies have suggested that subunits may possess action independent of CaN activity (Kayyali et al., 1997; Asai et al., 1999; Zhou et al., 1999). The CnA subunit can become constitutively active if proteolytic cleavage removes the autoinhibitory site (Manalan and Klee, 1983), and the expression of the CnAα isoform has been shown to be important to LTP and synaptic plasticity of the CA1 and CA1–2 pyramidal neurons (Victor et al., 1995; Zhuo et al., 1999; Zeng et al., 2001; Groth et al., 2003). Following TBI there is a chronic failure in LTP induction (Sanders et al., 2000) that can be ameliorated with CaN inhibition (Albensi et al., 2000). Alterations in CaN subunit distribution may provide a potential mechanism for this loss of synaptic plasticity. Following TBI there is a persistent dysfunction in learning and memory (Dixon et al., 1996; Hamm et al., 1996; Dixon et al.,1997; Scheff et al., 1997; Kline et al., 2000, 2002). Alterations in LTP (Reeves et al., 1995) and LTD (Albensi et al., 2000) following TBI could be in part due to rapid, persistent alterations in CaN A subunit dendritic distribution as demonstrated in this study.
Differences in CnAα distribution alterations compared to CnAβ have implications for isoform specific actions following TBI. CnAβ staining appears to be primarily localized to dendrites within the SR of the CA1 and CA1-2 and within the exposed blade of the dentate suggesting that CnAβ function may be more important to plastic events relevant to synaptic connections within these regions. Prior developmental studies have suggested that both CnAα and CnAβ have important roles in synaptic organization and synaptogenesis (Eastwood et al., 2005).
There were decreases in IR at both 2h and 2 weeks in both isoforms within the ipsilateral exposed blade of the DG (Figs. 3–6). Given the projections of the DG through CA3 and into the CA1 region, decreases in isoform IR may not be due to acute alterations in hippocampal regions, but rather may be due to a loss of CaN isoform expression throughout projecting neurons. This would be consistent with recent experimental results demonstrating significant axonal loss throughout the gray matter of the hippocampus (Hall et al., 2008). Losses in projecting dendrites and subsequent alterations in synaptic targets would also provide a possible explanation for the reduction of CnA isoform IR within the SR of the CA1 and CA1–2 regions, given that a loss of synaptic input would preclude the necessity for CaN activity at synapses. This may indicate that there is a change in the pattern of CaN distribution (i.e., a movement from synapses to cell bodies) within these areas that may provide an explanation for why no gross CnA isoform protein expression differences can be appreciated with Western blot (Fig. 7).
In summary, CnA isoforms are differentially modulated by acute injury in a regionally-specific manner. Regional alterations in CnA isoforms within the hippocampus ipsilateral to injury persist chronically. Further studies are needed to examine exactly how isoform changes relate to CaN activity in different regions of the hippocampus. It is also unclear what the functional implications are of specific subunit isoform alterations. While some studies have shown cell death and LTP associations with CnAα, it remains unclear what the CnA subunit's function is independent of the activity of the CaN heterodimer. These regionally-specific alterations in isoform expression also have important implications for the clinical use of CaN inhibitors in TBI. This study is the first step in identifying these alterations and characterizing specific CaN subunit changes following TBI that will allow for more precisely targeted therapeutic strategies to facilitate recovery from TBI.
We would like to thank Melanie Grubisha for her invaluable advice regarding this research.
No conflicting financial interests exist.