Temporal-Specific Sex and Injury-Dependent Changes on Neurogranin-Associated Synaptic Signaling After Controlled Cortical Impact in Rats

Abstract

Extensive effort has been made to study the role of synaptic deficits in cognitive impairment after traumatic brain injury (TBI). Neurogranin (Ng) is a calcium-sensitive calmodulin (CaM)–binding protein essential for Ca²⁺/CaM-dependent kinase II (CaMKII) autophosphorylation which subsequently modulates synaptic plasticity. Given the loss of Ng expression after injury, additional research is warranted to discern changes in hippocampal post-synaptic signaling after TBI. Under isoflurane anesthesia, adult, male and female Sprague–Dawley rats received a sham/control or controlled cortical impact (CCI) injury. Ipsilateral hippocampal synaptosomes were isolated at 24 h and 1, 2, and 4 weeks post-injury, and western blot was used to evaluate protein expression of Ng-associated signaling proteins. Non-parametric Mann–Whitney tests were used to determine significance of injury for each sex at each time point. There were significant changes in the hippocampal synaptic expression of Ng and associated synaptic proteins such as phosphorylated Ng, CaMKII, and CaM up to 4 weeks post-CCI, demonstrating TBI alters hippocampal post-synaptic signaling. This study furthers our understanding of mechanisms of cognitive dysfunction within the synapse sub-acutely after TBI.

Introduction

A pervasive and persistent challenge faced by traumatic brain injury (TBI) survivors is a significant disturbance in cognitive function. Learning, memory, and attention faculties are especially vulnerable across the spectrum of injury severity and symptoms may persist for years to decades [1–7]. Several mechanisms have been posited as contributors of cognitive deficits in TBI. These include neuronal axonal injury, excitotoxicity, oxidative stress, mitochondrial mal-function, inflammation, and various forms of programmed cell death. Given that the synapse is considered the functional unit of the brain [8], extensive effort has been made to study the role of synaptic plasticity deficits in persistent cognitive function after TBI. Synaptic deficits have been associated with reduced axoplasmic transport caused by axonal injury [9], synaptic mitochondrial impairments [10], oxidative stress modulation by synapse loss and replacement [11], and synaptic remodeling by neuroinflammatory responses to TBI [12]. Therefore, synaptic plasticity deficits may be at the nexus of multiple mechanisms of TBI-mediated cognitive dysfunction.

Several studies have investigated and characterized structural and electrophysiological changes in synaptic plasticity after experimental TBI in rodents. Reductions in hippocampal synaptic count and dendritic spine maturity were observed days to months after injury [13–17]. These neuroanatomical changes are accompanied by diminished hippocampal long-term potentiation (LTP) at similar time points [18–26]. Biochemical underpinning of learning and memory is facilitated by activation of multiple kinases, such as extracellular signal-regulated kinases (ERK), calcium/calmodulin-dependent kinases (CaMK), and protein kinase C (PKC) that subsequently modulate multiple signaling pathways in synapses of the hippocampus [27–33]. After experimental TBI, hippocampal changes in kinase activity and expression and bioavailability of downstream substrates are modulated acutely after injury in response to excitotoxicity. However, many of these changes have been shown to recover within days of the injury [34–39], or measurements have not been documented beyond acute time points. Given behavioral deficits in pre-clinical models have been observed up to 1 year after experimental injury [40, 41], identifying underlying molecular changes in the subacute to chronic phases of TBI are key to understanding potential rehabilitation strategies.

Neurogranin (Ng) is a small, 7.6kD, post-synaptic protein that plays an essential role in enhancing synaptic strength by fine-tuning LTP [42, 43]. Studies describe Ng as a calmodulin (CaM) shuttle, localizing and releasing CaM in a precisely timed matter for downstream signaling [44, 45] (Fig. 1a). Increased post-synaptic calcium levels by neuronal stimulation cause two Ng-specific processes to occur. First, Ng-bound CaM is released from this complex, thereby allowing CaM to bind to co-localized CaMKII [46]. This interaction activates CaMKII, initiating autophosphorylation which is necessary for the induction and maintenance of synaptic plasticity [30, 47–52]. Constitutive transgenic knock-down of Ng demonstrates lower autophosphorylated CaMKII (P-CaMKII) and worsened performance in spatial learning tasks [43, 53, 54]. Second, as Ng becomes unbound from CaM, Ng is phosphorylated (P-Ng) by Ca²⁺-activated PKC, thereby prohibiting Ng’s re-binding with CaM [55–58]. Ng phosphorylation and dephosphorylation are critical for fine-tuning synaptic plasticity as phospho-inhibiting and phospho-mimicking mutations at the Serine 36 phosphorylation site result in differentially impaired synaptic communication [42].

Fig. 1.

Schematic of Ng signaling pathway and experimental design. A Ng-associated signaling pathway demonstrates how Ng interactions with CaM lead to phosphorylation of Ng and CaMKII after calcium influx. B Illustration shows experimental groups and design. Male and female Sprague–Dawley rats received either sham or controlled cortical impact (CCI) injury at ipsilateral hippocampus was collected at indicated time points for western blot (WB) for indicated proteins. Mann–Whitney non-parametric statistical analysis was conducted between sham and CCI groups of each sex at each time point. Ng neurogranin, CaM calmodulin, Ca²⁺ calcium, CaMKII calcium/calmodulin-dependent kinase II, PKC protein kinase C, P-Ng Ng phosphorylated at Ser36, P-CaMKII calcium/calmodulin-dependent kinase II phosphorylated at Thr 286, WB western blot

Ng has recently been identified as a candidate fluid biomarker of cognitive decline across multiple neurodegenerative diseases, including TBI [59]. Specifically, studies have shown significantly increased levels of serum Ng acutely in patients with mild TBI, alongside a reduction of Ng levels in exosomes of military personnel months after blast exposure [60–62]. Furthermore, our group was the first to observe decreases in whole-cell hippocampal Ng expression up to 2 weeks after controlled cortical impact (CCI) [63]. This study aims to investigate the effect of CCI on Ng-associated hippocampal post-synaptic signaling pathway that may contribute to learning and memory deficits after TBI. We hypothesize that TBI decreases hippocampal synaptic expression of Ng and disrupts downstream synaptic P-CaMKII and P-Ng expression. To enhance our assessment of synaptic Ng signaling changes, we employed a strategy to test these protein changes in synaptosomal isolated hippocampal lysates. To this end, we conducted a time course examining synaptic ipsilateral hippocampal protein expression of Ng and associated signaling proteins P-Ng, α-CaMKII, P-CaMKII, and CaM from 24 h to 4 weeks after CCI injury in male and female animals. Post-synaptic density protein 95 (PSD-95) was added as a canonical marker of synaptic density for comparison [64, 65].

Methods

Animal Care and Controlled Cortical Impact (CCI)

All experimental procedures were approved by the University of Pittsburgh Institutional Animal Care and Use Committee in accordance with the guidelines established by the National Institutes of Health in the Guide for the Care and Use of Laboratory Animals. Animals were housed up to two rats per cage in the University of Pittsburgh vivarium with a 12:12 light/dark photoperiod (lights on at 7:00 a.m.) and provided food and water ad libitum. Animals were also monitored daily by veterinary technicians.

A total of 48 male and 48 female adult, Sprague–Dawley rats (250–350 g, Envigo, Indianapolis, IN) were used for this study, with 6 animals per injury group per time point (Fig. 1b). Group sizes were determined based on observed injury effects from a previous study with a similar experimental design [63]. In female rats, vaginal lavage and estrous cycle determination was conducted once immediately prior to surgical procedures, using methods previously described [66, 67]. Animals were initially anesthetized with 4% isoflurane in 2:1 N₂O/O₂ and maintained using 2% isoflurane in 2:1 N₂O/O₂. Following intubation, rats were placed on a homeothermic blanket (Harvard Apparatus, Holliston, MA) to regulate and maintain body temperature (37 °C). The animal’s head was placed in a stereotaxic frame (Kopf Instruments, Tujunga, CA) and a 7 mm craniectomy was performed over the right parietal cortex, between bregma and lambda, and centered 5 mm lateral of the sagittal suture to expose the dura mater to allow access for the impactor tip (6 mm flat tip) of the CCI device (Pittsburgh Precision Instruments Inc., Pittsburgh PA), as previously described [68, 69]. CCI at a depth of 2.5 mm at 4 m/s, with a dwell time of 150 ms, was carried out. After injury, the incision site was closed by silk sutures, and animal recovery was monitored by measuring time to reach righting reflex. The righting reflex was also used to monitor injury severity [70]. Sham (control) injury animals were subjected to identical anesthesia and surgical procedures but did not receive a TBI. One sham and one CCI animal died immediately following surgical procedures in the 1-week female cohort. One 24 h female animal was excluded from all post-mortem outcome measures for seizure-like behaviors. Lastly, one male animal was not included in CaM western blots at the 4-week time point (Fig. 6) due to limited sample availability.

Fig. 6.

Synaptic CaM expression is reduced at only 1 week post-injury. A Synaptosomal protein expression of CaM (17kD) and actin (42kD) by SDS-PAGE western blot in male and female animals after 24 h to 4 weeks after sham or CCI injury. Representative n = 2 animals per group are shown (N = 6 total animals per group). B Bar graphs show quantification of CaM expression normalized actin. Data is represented as mean ± standard error of the mean and normalized to corresponding sham group. Mann–Whitney test was used to analyze injury effect for each sex, at each time point. *p < 0.05, **p < 0.01, compared to sham animals

Synaptosome Isolation

At 24 h or 1, 2, or 4 weeks post-injury, animals received an overdose of sodium pentobarbital (intraperitoneally, 100 mg/kg Fatal-plus, Vortech Pharmaceuticals, Dearborn, MI) and were rapidly decapitated. The ipsilateral hippocampus was rapidly dissected on a chilled ice plate and immediately frozen in liquid nitrogen and stored at − 80 °C. Tissue was homogenized using a dounce homogenizer in Syn-PER lysis buffer (Invitrogen). Lysates were centrifuged at 1200 × g for 10 min at 4 °C. The supernatant was collected and centrifuged at 15,000 × g for 20 min at 4 °C and the synaptosomal pellet was diluted in Syn-PER buffer, as we previously described [68, 71]. Protein concentration was measured with a bicinchoninic acid protein assay kit (Thermo Scientific, Pittsburgh PA) using a 96-well microplate reader (Biotek, Winooski, VT).

Western Blotting

To assess synaptic expression of Ng-associated signaling proteins, synaptosomal lysates were boiled for 5 min prior to blotting. Fifteen micrograms of ipsilateral hippocampal protein samples and molecular weight markers (Bio-Rad, Hercules, CA) were separated using sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). The resolved proteins were electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane (Invitrogen, Carlsbad, CA). The blots were blocked in 5% nonfat dry milk in tris-buffered saline (TBS) for 1 h. The primary antibody was diluted in a blocking solution and incubated overnight at 4 °C. The following day, the membranes were washed with 1 × TBS buffer, incubated in a blocking solution containing horseradish peroxidase secondary antibodies for 1 h. Proteins were visualized using a chemiluminescence detection system (Supersignal, Pierce). The membranes were stripped and probed with an anti-β-actin antibody. Blots were imaged with the Chemidoc Imager (Bio-Rad). Optical densities were measured using ImageJ (National Institutes of Health) and normalized to β-actin levels. Values are presented as the ratio of optical densities of samples as a percentage of sham (100%) for each sex for each time point.

Antibodies

The following antibodies were used in described experiments: rabbit anti-neurogranin (1:4000, Abcam, ab217672) as previously described [63, 68], rabbit anti-phosphorylated neurogranin (Ser36, 1:1000, Millipore, ABN426), rabbit anti-PSD-95 (1:1000, Abcam, ab18258), rabbit anti-phosphorylated CaMKII (Thr286, 1:5000, Cell Signaling, 12,716), mouse anti-α-CaMKII (1:2000, Cell Signaling, 50,049), rabbit anti-Calmodulin (1:1000, Abcam, ab45689), and mouse anti-β-actin (1:2000, Sigma, A5316).

Statistics

Values for protein abundances determined by western blot are presented as the ratio of optical densities of samples as a percentage of sham (100%) for each sex at each time point. Both P-Ng and P-CaMKII expression were normalized to total protein expression of α-CaMKII and Ng. Data are presented as bar graphs with individual values showing the mean ± standard error of the mean (SEM). Animal weight and righting time were evaluated by two-way ANOVA to determine the main effects of injury, sex, and injury × sex interactions. For immunoblot data, Sham and CCI comparisons for each sex at each time point were conducted using the Mann–Whitney non-parametric test. A p-value p < 0.05 was considered statistically significant for all tests. Qualitative observations of sex differences were noted. Statistical tests were completed using GraphPad Prism (GraphPad, La Jolla, CA). Whenever possible, investigators were blinded to injury group assignment, particularly during tissue processing, experimentation, and data analysis.

Results

Injury and Sex-Dependent Observations on Animal Weights, Righting Times, and Estrous Cycle

Animal weight and right time were measured in all animals immediately prior to or after surgical procedures, respectively, and two-way ANOVA was conducted to examine the effects of injury and sex (Fig. 2). For animal weight, there was a significant main effect of sex (p < 0.0001), with no significant effect of injury and no significant injury × sex interaction (Fig. 2a). For righting time, there were significant main effects of injury (p < 0.0001), sex (p = 0.0224), and injury × sex interaction (p = 0.0488) (Fig. 2b). Post hoc showed CCI-male animals had significantly longer right time than CCI-female animals (p = 0.014), while there were no differences in righting times between sham-male sham-female groups. The estrous cycle was determined in female animals only immediately prior to surgical procedures. Chi-square test showed no significant differences between estrous phases between female animals between the sham and CCI groups (Fig. 2c).

Fig. 2.

Weights, righting times, and estrous cycle at time of injury. A Animal weights were recorded prior to sham or CCI injury. Data is represented as mean ± standard error of the mean. Two-way ANOVA was used to analyze injury and sex effects. B Righting times were recorded immediately after sham and CCI surgical procedures. Data is represented as mean ± standard error of the mean. Two-way ANOVA was used to analyze injury and sex effects. C Estrous cycle was measured in female animals prior to sham or CCI injury. Chi-square analysis shows no significant differences in phases of estrous cycle between sham and CCI female animals. ****p < 0.0001 compared to sham (overall injury effects); &p < 0.05, &&p < 0.01 compared to male (overall sex effects)

Synaptic Hippocampal Ng Expression Is Reduced up to 4 Weeks Post-injury

Changes in synaptic hippocampal expression of Ng were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 3b, c). At 24 h post-injury, there was no significant difference in Ng expression between male and female sham and CCI animals, although female CCI animals showed a trending decrease (p = 0.0519). At 1 week post-injury, male and female CCI animals showed a significant decrease in Ng expression compared to male and female sham animals (p = 0.0043 and p = 0.0043, respectively). At 2 and 4 weeks post-injury, male CCI animals showed a trending decrease in Ng expression compared to male sham animals (p = 0.0584 and p = 0.0931, respectively). At 2 and 4 weeks post-injury, female CCI animals showed a significant decrease in Ng expression compared to female sham animals (p = 0.0397 and p = 0.0260, respectively).

Fig. 3.

Synaptic hippocampal Ng expression is reduced up to 4 weeks post-injury. A Synaptosomal protein expression of Ng (15kD) and actin (42kD) by SDS-PAGE western blot in male and female animals after 24 h to 4 weeks after sham or CCI injury. Representative n = 2 animals per group are shown (N = 6 total animals per group). B Bar graphs show quantification of Ng expression normalized actin. Data is represented as mean ± standard error of the mean and normalized to corresponding sham group. Mann–Whitney test was used to analyze injury effect for each sex, at each time point. *p < 0.05, **p < 0.01, compared to sham animals

Synaptic Hippocampal P-Ng Ratio Demonstrates Bi-phasic Injury and Sex Differences

Changes in P-Ng synaptic hippocampal expression were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 4b, c). At 24 h post-injury, there was no significant difference in P-Ng expression between sham and CCI for males and females. At 1 and 2 weeks post-injury, P-Ng expression was not significantly different between male sham and CCI animals. At 1 and 2 weeks post-injury, female CCI animals showed a significant decrease in P-Ng expression compared to female sham animals (p = 0.0022 and p = 0.0397, respectively). At 4 weeks post-injury, there was no significant difference in P-Ng expression between male and female sham and CCI animals, although male CCI animals showed a trending increase (p = 0.0649).

Fig. 4.

Synaptic hippocampal P-Ng ratio demonstrates bi-phasic injury and sex differences. A Synaptosomal protein expression of P-Ng (15kD) and actin (42kD) by SDS-PAGE western blot in male and female animals after 24 h to 4 weeks after sham or CCI injury. Representative n = 2 animals per group are shown (N = 6 total animals per group). Bar graphs show quantification of B P-Ng expression normalized actin and C P-Ng normalized to total Ng expression, to show P-g/Ng ratio. Data is represented as mean ± standard error of the mean and normalized to corresponding sham group. Mann–Whitney test was used to analyze injury effect for each sex, at each time point. *p < 0.05, **p < 0.01, compared to sham animals

P-Ng synaptic expression was then normalized to total protein to determine a P-Ng/Ng ratio (Fig. 4d, e), to account for CCI-induced changes in synaptic Ng expression. Changes in P-Ng/Ng ratio were analyzed by the Mann–Whitney test for each sex and time point. At 24 h post-injury, there was no significant difference in P-Ng/Ng ratio between sham and CCI for males and females. At 1 week post-injury, P-Ng/Ng ratio was not significantly different between male sham and CCI animals. At 1 week post-injury, female CCI animals showed a significant increase in P-Ng/Ng ratio compared to female sham animals (p = 0.0238). At 2 weeks post-injury, there was no significant difference in P-Ng/Ng ratio between sham and CCI for males and females although female animals show a trending decrease (p = 0.0556). At 4 weeks post-injury, P-Ng/Ng ratio was significantly increased in both male and female CCI animals compared to male and female sham animals (p = 0.0152 and p = 0.0260, respectively).

Synaptic Hippocampal CaMKII Expression Demonstrates Injury and Sex Differences

Changes in P-CaMKII synaptic hippocampal expression were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 5b, c). At 24 h post-injury, there was no significant difference in P-CaMKII expression between sham and CCI for males and females, although female CCI animals showed a trending increase (p = 0.0931). At 1 week post-injury, male CCI animals showed a significant decrease in P-CaMKII expression compared to male sham animals (p = 0.0260). At 1 week post-injury, P-CaMKII expression was not significantly different between female Sham and CCI animals. At 2 weeks post-injury, P-CaMKII expression was not significantly different between male sham and CCI animals. At 2 weeks post-injury, female CCI animals showed a significant decrease in P-CaMKII expression compared to female sham animals (p = 0.0079). At 4 weeks post-injury, there was no significant difference in P-CaMKII expression between male and female sham and CCI animals.

Fig. 5.

Synaptic hippocampal CaMKII expression demonstrates injury and sex differences. A Synaptosomal protein expression of P-CaMKII (50, 60kD), α-CaMKII (50kD), and actin (42kD) by SDS-PAGE western blot in male and female animals after 24 h to 4 weeks after sham or CCI injury. Representative n = 2 animals per group are shown (N = 6 total animals per group). Bar graphs show quantification of B P-CaMKII expression normalized actin, C α-CaMKII expression normalized actin, and D P-CaMKII normalized to α-CaMKII expression to show P-CaMKII/α-CaMKII ratio. Data is represented as mean ± standard error of the mean and normalized to corresponding sham group. Mann–Whitney test was used to analyze injury effect for each sex, at each time point. *p < 0.05, **p < 0.01, compared to sham animals

Changes in α-CaMKII synaptic hippocampal expression were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 5d, e). At 24 h post-injury, there was no significant difference in α-CaMKII expression between sham and CCI for males and females, although female CCI animals showed a trending increase (p = 0.0823). At 1 and 2 weeks post-injury, male and female CCI animals showed a significant decrease in α-CaMKII expression compared to male and female sham animals (male: p = 0.0411 and p = 0.0216, respectively; female: p = 0.0022 and p = 0.0079, respectively). At 4 weeks post-injury, there was no significant difference in α-CaMKII expression between male and female sham and CCI animals,

P-CaMKII synaptic expression was then normalized to α-CaMKII expression to determine changes in the P-CaMKII/α-CaMKII ratio. Changes in P-CaMKII/α-CaMKII ratio were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 5f, g). At 24 h post-injury, male CCI animals showed a significantly higher P-CaMKII/α-CaMKII ratio than male sham animals (p = 0.0260). At 24 h post-injury, there was no significant difference in P-CaMKII/α-CaMKII ratio between female sham and CCI animals. At 1, 2, and 4 weeks post-injury, there was no significant difference in P-CaMKII/α-CaMKII ratio between male and female sham and CCI animals.

Synaptic Hippocampal CaM Expression Is Decreased Only at 1 Week Post-injury

Changes in CaM synaptic hippocampal expression were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 6b, c). At 24 h post-injury, there was no significant difference in CaM expression between male and female sham and CCI animals. At 1 week post-injury, male and female CCI animals showed a significant decrease in CaM expression compared to male and female sham animals (p = 0.0476 and p = 0.0087, respectively). At 2 and 4 weeks post-injury, there was no significant difference in CaM expression between male and female sham and CCI animals.

Synaptic Hippocampal Changes in PSD-95 Expression Recovers by 2 Weeks Post-injury

Changes in PSD-95 synaptic hippocampal expression were analyzed by the Mann–Whitney test for each sex at each time point (Fig. 7b, c). At 24 h post-injury, male and female CCI animals showed a significantly increased PSD-95 expression than male and female sham animals (p = 0.0087 and p = 0.0087, respectively). At 1 week post-injury, male and female CCI animals showed a significantly decreased PSD-95 expression than male and female sham animals (p = 0.0087 and p = 0.0022, respectively). At 2 and 4 weeks post-injury, there was no significant difference in PSD-95 expression between male and female sham and CCI animals, although female animals showed a trending decrease (p = 0.0952 and p = 0.0931, respectively).

Fig. 7.

Synaptic hippocampal changes in PSD-95 expression recovers by 2 weeks post-injury. A Synaptosomal protein expression of PSD-95 (95kD) and actin (42kD) by SDS-PAGE western blot in male and female animals after 24 h to 4 weeks after sham or CCI injury. Representative n = 2 animals per group are shown (N = 6 total animals per group). B Bar graphs show quantification of PSD-95 expression normalized actin. Data is represented as mean ± standard error of the mean and normalized to corresponding sham group. Mann–Whitney test was used to analyze injury effect for each sex, at each time point. **p < 0.01, compared to sham animals

Discussion

We found that CCI significantly decreases hippocampal synaptic Ng expression in female animals 1 to 4 weeks post-CCI, while Ng expression in male animals recovers significantly by 2 weeks post-CCI. P-Ng/Ng ratio showed bi-phasic increases post-CCI. Male animals demonstrated an increase in P-CaMKII/α-CaMKII ratio 24 h post-CCI. CaM and PSD-95 expression was altered early after injury but was recovered by 2 weeks post-CCI. There is evidence to support injury and sex-dependent changes in expression and phosphorylation state of Ng-associated signaling proteins across various time points after CCI. This is the first study to examine synaptic expression of Ng and associated proteins in both male and female animals up to the 4-week post-injury time point.

Examining a time course allows for temporal resolution of synaptic signaling changes in the context of known pathological and rehabilitative injury processes at similar times. At 1 week post-injury, we observed decreases in protein expression across most proteins measured, perhaps reflective of hallmark cellular and synaptic loss observed after experimental TBI. Previous studies examining synapse loss and reduction in synaptic density show decreases in both pre- and post-synaptic markers across the hippocampus up to 7 days after CCI [14, 72, 73]. Specifically, PSD-95 expression is lower at 7 days post-injury [63, 74], a finding replicated in this study. By the 2-week time point, we observe a divergence in the recovery of synaptic protein expression. While CaM expression is recovered by 2 weeks, P-CaMKII and α-CaMKII levels return to sham levels by 4 weeks post-injury and Ng expression does not recover through 4 weeks. This suggests that changes observed at the 2 and 4-week time points occur independently of synaptic loss and endogenous recovery mechanisms alone cannot restore plasticity-related processes.

The 24 h and 4-week time points showed different outcomes in Ng synaptosomal expression to our previously published assessment of TBI-induced changes in whole-cell Ng expression. Normally, Ng mRNA is transported to the synapse where it is locally translated in a highly dynamic and calcium-dependent manner [75]. At 24 h post-injury, decreased Ng expression was observed in whole cell samples, while synaptic expression was unchanged. This may be due to excitotoxicity observed acutely after CCI, with higher levels of calcium being localized to the synaptic areas, compared to the entire neuron [76]. Therefore, local translation of Ng mRNA in the synapse may offset overall decreases in Ng expression across the neuron. Conversely, at 4 weeks post-injury, whole cell expression of Ng was restored while synaptic expression was decreased. This may be due to impairments in Ng mRNA transport to the synapse due to axonal injury or impaired translation within the synapse [77–80]. This highlights the importance of incorporating methods to examine sub-cellular localization as it may elucidate underlying mechanisms of changes in protein expression after TBI.

Phosphorylation of Ng has been shown to be an important regulator of synaptic plasticity and disruption of this process may have implications to cognitive outcomes [42, 45, 56]. There is a bi-phasic increase in the P-Ng/Ng ratio at the 1 and 4-week time points. Factors contributing to these increases may diverge given injury-related processes occurring at this time point. At 1 week, there is a significant decrease in P-Ng expression in female CCI animals compared to female sham animals in contrast to the 4-week time point where we observed a trending increase in P-Ng expression in male CCI animals compared to male sham animals. This distinction could be important in the context of the state of the synapse between these two time points and reflective of synaptic damage and recovery. Hippocampal expression of P-Ng was increased immediately in models of progressive auto-hypoxia and also in electroconvulsive seizure, reflective of acute injury response [81, 82]. Ng is phosphorylated by PKC and de-phosphorylated by calcineurin [55, 58, 83]. Although hippocampal PKC expression and activity is increased minutes to a few hours after fluid percussion injury, we do not observe increased P-Ng expression at the 24-h time point with the CCI model. Examination of P-Ng expression in the minutes to hours after injury may show tighter association between PKC activity and its substrate, Ng. Increased P-Ng was identified in the cerebrospinal fluid of human TBI patients by mass spectrometry within 10 days of injury [84]. Whether the observed increase in P-Ng/Ng ratio at the 4-week time point is pathological or compensatory is unknown. It has been previously shown that phospho-inhibiting mutations of Ng do not interfere with CaMKII-mediated signaling, while phospho-mimicking mutants of Ng showed this form completely blocks the induction of LTP [42]. Although LTP is not completely abolished after CCI at this time point [24], this data suggests that increased P-Ng/Ng ratio at 4 weeks may contribute to dysfunctional synaptic outcomes.

While previous studies showed a strong relationship between Ng expression and CaMKII autophosphorylation [53], we did not see decreases in P-CaMKII/α-CaMKII ratio at time points when Ng was significantly decreased. Although decreases in P-CaMKII expression alone were observed, these appear to be attributed to decreased overall loss in α-CaMKII expression. Similarly, a recent study knocked down Ng expression in cultured hippocampal neurons and saw no change in expression of P-CaMKII expression, while electrophysiological metrics of synaptic plasticity were disrupted [85]. They postulated that assessment of P-CaMKII by western blot could be measuring baseline levels of protein expression and not changes due to neuronal stimulation upon which changes in Ng expression could show an effect. This corresponds to previous studies examining P-CaMKII/α-CaMKII ratio across multiple experimental TBI models reporting injury-induced disruption of calcium homeostasis initially increases P-CaMKII minutes after TBI, with recovery to baseline by 24 h [36, 37, 86–88]. Similarly, we observed a CCI-dependent increase in P-CaMKII//α-CaMKII ratio in male animals and an increase in PSD-95 expression in both sexes at our most acute time point of 24 h, suggesting lingering effects of these acute secondary injury processes. Future studies incorporating a stimulation paradigm, either through examinations immediately after a behavioral paradigm or electrophysiological stimulation, may elucidate changes in P-CaMKII expression after TBI [75, 89].

Alternatively, Hwang et al. also postulate that decreases in Ng expression increase free CaM to sequester calcium, reducing overall calcium levels. This subsequently activates calcineurin, a phosphatase that modulates synaptic strength. We observe not only decreases in synaptic Ng, but also increases in P-Ng expression, a state that does not allow Ng to bind to CaM, which compounds this phenomenon further. While calcineurin expression is decreased and redistributed after experimental TBI [90–92], increased calcineurin activity has been shown up to 4 weeks after experimental TBI [16]. Calcineurin inhibitors have been evaluated as a potential therapeutic intervention after TBI [26, 93–95]. Further studies examining the relationship between Ng expression and calcineurin activity can elucidate this alternative pathway of Ng-mediated synaptic dysfunction after TBI.

Interestingly, we observed qualitative sex differences in response to CCI injury across protein and time points. Male CCI animals appear to have faster recovery of synaptic hippocampal Ng expression than female CCI animals. Female CCI animals also show a bi-phasic increase in P-Ng/Ng ratio, while male CCI animals only demonstrate this increase at 4 weeks post-injury. Lastly, P-CaMKII/α-CaMKII ratio is only increased at 24 h post-injury in male CCI animals. Our righting reflex data shows CCI-injured male animals have significantly longer righting times than CCI-injured female animals, suggesting female animals experienced a slightly less severe injury than male animals. This is in line with previous literature showing female animals have smaller lesion volumes and less edema than male animals [96]. A review of experimental TBI studies examining sex differences in outcomes reports male animals more often have worse outcomes after CCI compared to female animals [97]. While studies examining sex differences specifically within the synapse after TBI are limited, Semple et al. showed that 3 weeks after pediatric TBI, male animals show significantly reduced hippocampal and cortical dendritic morphology from their sham counterparts, while injured females showed no injury effect [98]. This was associated with males demonstrating worsened performance in select cognitive tasks. Similarly, clinical TBI studies show men have worse outcomes, including performance in memory tasks, than women after moderate-severe TBI, particularly in studies with over 1000 participants [94]. However, without corresponding behavioral data in the current dataset, we are limited in discerning whether sex differences observed in this study may be indicators of potential contributors to differences in outcome post-injury. Continued incorporation of both sexes in future examinations of synaptic changes after TBI can aid in elucidating mechanisms of dysfunction and therapeutic strategy.

There are a few limitations to this study. Given the exploratory nature of the study, no formal sample size calculations were done before the onset of the experiments and no systematic blinding was implemented. The group sizes of 5–6 per group were decided based on the results from a previous investigation of whole cell Ng and PSD-95 expression [63]. However, the use of multiple Mann–Whitney non-parametric tests leads to increases in type 1 error, where the null hypothesis is incorrectly rejected. Given the number of trending differences between sham and CCI groups across our proteins, replication of this study with larger group sizes is needed. Along those lines, a previous study from our group showed a significant difference in synaptic hippocampal expression of Ng at 2 weeks between sham-Veh and CCI-Veh with group sizes of 9–10 animals [68]. Although we were able to observe some interesting qualitative sex differences in synaptic protein expression after CCI, due to experimental design, no statistical comparison of sex differences could be made. Running both male and female animals on the same gel would allow for statistical comparisons of sex differences. Lastly, as mentioned above, this study is limited by the lack of functional outcomes in association with observed molecular changes. Although investigation of expression of proteins involved in synaptic plasticity in behaviorally naïve animals is a crucial first step, concurrent behavioral evaluation allows us to elucidate these changes in the context of post-TBI cognitive deficits.

In conclusion, CCI significantly reduces hippocampal synaptic Ng expression and significantly alters expression of Ng-associated signaling proteins in a temporal and sex-specific manner. Identifying pathological synaptic signaling processes or potential compensatory mechanisms that correspond to more chronic cognitive symptoms after TBI can help direct future therapeutic and rehabilitation strategies.

Data Availability

Data will be made available on request.