INTRODUCTION
Traumatic brain injury (TBI) is a permanent or temporary neurological dysfunction caused by an external force to the brain. It is one of the most common causes of neurological sequela, but little improvement in the mortality and chronic disability rates caused by TBI has been reported over the last 20 years [1]. TBI can impair physical, cognitive, behavioral, and social functions [2]. While the long-term prospects for mild to moderate cognitive recovery following TBI are generally positive, some patients will experience aftereffects [2,3].
Biomarkers of TBI can suggest suitable treatment methods to reduce long-term effects and secondary brain damage. The use of TBI biomarkers enhances clinical prognosis, aids in the formulation of treatment strategies, and diminishes the incidence of severe outcomes [4].
Ubiquitin C-terminal hydrolase-L1 (UCH-L1) protein is a potential candidate biomarker for TBI. Due to its high and specific expression in neurons, this protein was previously used as a histological marker. UCH-L1 plays a role in the addition or removal of ubiquitin from proteins involved in metabolism [5]. It is crucial for elimination of excess, oxidized, or misfolded proteins in neurons under both healthy and neuropathological conditions [6]. An increase in UCH-L1 levels in cerebrospinal fluid (CSF) has been detected in patients with subarachnoid hemorrhage and severe TBI [7].
Glial fibrillary acidic protein (GFAP) is a monomeric intermediate filament predominantly located in the astroglial cytoskeleton. It is exclusive to brain tissue and does not typically circulate in peripheral blood. Astrocytes undergoing cell death release GFAP, which is an ideal candidate marker for TBI [8]. According to recent studies, GFAP has superior prognostic value to other biomarkers in patients with TBI [9]. Serum GFAP levels change over time, as does its predictive value, which remains unknown during the acute stage after the injury.
Inflammatory mediators such as cytokines, chemokines, cytotoxic proteases, and oxygen radicals are also released after TBI. Interleukin 6 (IL-6) is known to control disease, immunity, and neural development [10]. Although normal brain tissue and CSF rarely exhibit acute local and systemic release into brain tissue, IL-6 can be detected in response to an injury [11]. Although it is extremely sensitive to brain injury, because IL-6 is released after peripheral injuries, data on its predictive ability are insufficient, particularly in TBI patients who also have existing injuries [12]. IL-6 is essential for both the development of early focused treatments and for understanding the neurological consequences of TBI on systemic inflammatory responses [13,14].
This study investigates experimental brain trauma in rabbits, assessing UCH-L1, GFAP, and IL-6 levels in serum and CSF and comparing these biomarkers among trauma groups.
METHODS
Ethics statement
This study was approved by the Ethics Committee of Selcuk University Hospital, Konya, Turkey (No. 2016-41/2). All procedures were performed in accordance with the institutional and national guidelines for the care and use of laboratory animals. All procedures were performed in accordance with institutional and national guidelines for the care and use of laboratory animals.
Study design
The study was conducted at Department of Emergency Medicine, Faculty of Medicine, Selcuk University. Thirty New Zealand rabbits, both male and female and weighing from 1,700 to 3,000 g, were used for this study. Following simple random sampling, the rabbits were split into four groups: control group (n=6), mild-trauma group (n=8), moderate-trauma group (n=8), and severe-trauma group (n=8). All rabbits received an intramuscular injection of 50 mg/kg ketamine and 15 mg/kg xylazine hydrochloride (Rompun) for anesthesia, and the central auricular artery and marginal auricular vein were catheterized.
The trauma model was modified from the Marmarou impact acceleration framework [15], with the rabbits arranged in a prone position on a horizontal platform. The rabbits were sedated and their fur was shaved. A midline scalp incision was performed under sterile conditions, and the periosteum covering the vertex was scraped with a dissector. To induce diffuse damage and increase the cranial contact surface, a stainless-steel metal disk was placed on the vertex, between the coronal and lambdoid sutures. To cause trauma, metal weights weighing 200, 350, or 500 g were dropped onto the metal disk positioned on the heads through a tube 1 m in length. The weights fell onto the disk under the effect of gravity. The layers were appropriately closed in accordance with standard procedures after skin haemostasis. Macroscopic views of the trauma moment and specimen are summarized in Fig. 1.
Although the mechanical aspects of the Marmarou impact acceleration model were used in the study, we modified the method and/or energy levels. Unlike the original Marmarou model, instead of applying weights of the same mass from different heights to create the trauma, weights of different masses were dropped from the same height (1 m). To prevent secondary skull fractures from objects applied to the base of the skull, a flat, round stainless-steel disk was used. For the same purpose, Marmarou used a protective disk with an area of π/4 cm2. Given that the skull thickness of rabbits is generally 1 to 2 mm and that of rats is 0.5 to 1 mm, and that rabbits have higher masses than rats, a protective disk with an area of π/16 cm2 (P=force/area) was used to increase the applied pressure force. In the evaluations performed, energy levels were defined (potential energy = mass × gravitational constant × height; kinetic energy = 1/2 × mass × velocity2), and impact velocities were 4.4 m/sec for all trauma models (gravitational constant ≈ 10 m/sec2 and π≈3.14 were used in the formulations, and friction forces were neglected). Energy levels were 2, 3.5, and 5 J for mild, moderate, and severe trauma, respectively. Because pressure transmitted to subjects at the time of impact is negatively correlated with surface area, the transmitted pressure forces were higher compared with those in the Marmarou model, and we concluded that the energy transmission of our model was sufficient. The rabbits were divided into three energy groups (mild, moderate, and severe) and parameter evaluations were compared. Energy levels and technical details regarding trauma groupings are summarized in Fig. 2.
To determine the levels of UCH-L1, GFAP, and IL-6, cerebrospinal fluid (CSF) samples were collected 0, 12, and 24 hours after head trauma. Each rabbit was positioned laterally with the head extended, and the occipitocervical region was shaved. A spinal needle was used to penetrate the atlantooccipital membrane and enter the intracisternal subarachnoid region after cleaning with an antiseptic solution. Approximately 0.1 mL of CSF was extracted within 3 minutes.
Electrocardiography and peripheral oxygen saturation (SPO2) monitoring were used to identify asystole, and rabbits that experienced apnea during the 24-hour observation period were euthanized. After 24 hours, surviving rabbits were euthanized with twice the initial amount of anesthetic. Gross pathological changes in their brains were examined and recorded, and brain tissue was obtained from the frontoparietal area and examined histopathologically.
Histopathological methods
To assess the appropriateness of the experimental model, histopathological analysis was conducted on rabbit brain tissue. The frontal lobes were extracted and preserved in a 10% formaldehyde solution. The frontal cortex was isolated, processed for standard histology (dehydration through graded ethanol, xylene fixation, and paraffin embedding), and sectioned into 5-µm slices using a rotary microtome. These sections were mounted onto poly-L-lysine coated slides. For general assessment, sections were deparaffinized, rehydrated, and stained with hematoxylin-eosin (H&E). To assess apoptosis, immunohistochemical staining was conducted using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay and primary antibodies for caspase-3 and caspase-8 (Fig. 3, Supplementary Figs. 1–3). For immunohistochemistry, sections were incubated overnight with primary caspase-3 and caspase-8 antibodies, followed by 3 hours of incubation with Histostain Plus (Thermo Fisher Scientific) after washing. Diaminobenzidine staining was used for color development. Nuclei were counterstained with Mayer’s hematoxylin solution to identify live cells, and slides were mounted with Entellan medium (Merck KGaA). Specimens were examined using an Olympus BX51 trinocular light microscope (Olympus Corp), and the average numbers of caspase-3, caspase-8, and TUNEL-labeled cells in the acquired images were quantified.
Image analysis
An Olympus BX51 trinocular microscope was used to examine the sections. An Olympus DP72 camera (Olympus Corp) was used to record digital images. Each tissue sample was inspected individually under the microscope, and three researchers examined the parameter at different times in a double-blind manner. The presence of hematoma, neuroinflammation (gliosis), increased vascularization, structural damage in neurons, edema, disruption of meningeal integrity, and axonal damage were defined (Fig. 3, Supplementary Figs. 1–3). A scoring scale designed for this evaluation was as follows: 0, absent; +, mild; ++, moderate; +++, severe; and ++++, extremely severe.
Biochemical methods
Blood and CSF samples were collected from all groups in gel tubes for measurement of GFAP, IL-6, and UCH-L1 levels at 0, 12, and 24 hours. Serum samples were centrifuged at 3,000 rpm for 10 minutes after waiting at least 10 minutes to allow clotting. The acquired serum samples were labeled, put in Eppendorf tubes, and kept at −80 °C until they could be analyzed.
After sample collection, all samples were removed from the freezer and thawed to room temperature for analysis. To ensure homogeneity, all samples were vortexed before analysis. The levels of GFAP in serum and CSF samples were measured using the GFAP Rabbit ELISA Kit 96 Test (ABBEXA ELISA kit, Abbexa). The levels of UCH-L1 were determined using the UCH-L1 Rabbit ELISA Kit 96 Test (MBS080655 ELISA kit, MyBioSource). IL-6 levels were measured using the IL-6 Rabbit ELISA Kit 96 Test (ABBEXA ELISA kit). All tests were performed using the enzyme-linked immunosorbent assays (ELISA) on the CLARIOstar ELISA reader (BMG LABTECH).
Statistical analysis
Our study set a target minimum power level of 80%. The required sample size for three repeated measures was calculated using G*Power (Heinrich Heine University Düsseldorf), with a type I error rate of 5% (α=0.05). In the analysis, the effect size (d) calculated by the direct method was 0.25, and the minimum required sample size for the study was 24 (critical F=2.73). Our study involved 30 rabbits, and a post hoc power analysis applying the same effect size and alpha values revealed a final power of 92.3%.
Statistical analysis for the study was conducted using IBM SPSS ver. 21 (IBM Corp). Because the quantitative data exhibited a non-normal distribution, nonparametric tests were carried out. Comparisons of repeated measures between multiple groups were carried out using a Friedman test. The Dunn-Bonferroni test was used for post hoc analysis. All statistical tests were conducted at a significance level of 0.05, ensuring that the results could be determined with a 95% confidence interval.
RESULTS
Histopathological analysis of brain tissue samples revealed a clear gradation of injury corresponding to the severity of the induced trauma. General histological assessment with H&E staining showed normal tissues in the control group and pathological changes in the trauma groups. Specifically, the moderate- and severe-trauma groups displayed notable disruptions of meningeal integrity, subpial hematomas, and gliosis. In addition, edema, increased vascularization, and structural damage in neurons demonstrated a progressive increase in prevalence from the mild- to the severe-trauma groups. An investigation of apoptosis also indicated a direct correlation between trauma severity and the rate of neuronal cell death. Immunohistochemical staining for caspase-3 and caspase-8 showed an increased number of positive neurons in the trauma groups, with the highest concentrations found in the severe-trauma cohort (Supplementary Figs. 1, 2). This trend was confirmed by the TUNEL assay, which also revealed progressively increasing numbers of apoptotic cells from the mild- to the severe-trauma group, supporting a link between trauma force and extent of apoptosis (Supplementary Fig. 3).
CSF biomarkers
The CSF levels of GFAP, IL-6, and UCH-L1 did not differ significantly between the mild-trauma group and control group (all P>0.05) (Table 1). The GFAP levels were highest in the control group, and the lowest level was observed at 0 hours in the moderate-trauma group (P=0.044). CSF levels of IL-6 and UCH-L1 level did not differ significantly between the groups (P>0.05) (Table 1).
In the severe-trauma group, the control group had the lowest GFAP levels, while the 12-hour sample after trauma showed the highest level of GFAP (P=0.002). Similarly, IL-6 levels peaked at 12 hours after trauma (P=0.014), although they were at their lowest in the control group. There was no significant difference in CSF UCH-L1 levels between the control and trauma groups (P=0.514). The CSF GFAP levels in the controls were significantly lower than those detected in the severe-trauma group at the 0-hour time mark, based on a pairwise comparison of all groups (P=0.003). At the 12-hour time point, CSF levels in the control group were significantly lower than those in the severe-trauma group (P=0.001). Additionally, the 24-hour CSF measurements were significantly higher compared with the control group (P=0.004). Regarding IL-6 levels, the control group had significantly lower CSF values compared with the severe-trauma group at 0 hours (P=0.001). The 12-hour CSF values of the severe-trauma group were significantly higher than the 0-hour CSF values of the group (P=0.010) (Table 1).
Serum biomarkers
Serum concentrations of GFAP, IL-6, and UCH-L1 did not differ significantly between the mild-trauma and control groups (all P>0.05) (Table 2). In the moderate-trauma group, serum IL-6 levels were lowest at 12 hours after trauma, while they were at the highest level after 24 hours (P=0.019). However, while serum UCH-L1 levels were at their lowest levels in the control group, they were at their highest levels at 12 hours after trauma (P=0.001). Although no significant difference was observed between the groups, serum GFAP levels were highest at 0 hours and tended to decrease progressively (P=0.517). When the pairwise comparisons between the groups for serum IL-6 were examined, the 12-hour serum IL-6 levels of the moderate-trauma group were significantly lower than those in the control group (P=0.004). In the pairwise comparisons of serum UCH-L1 levels between the groups, the moderate-trauma group had substantially higher 12-hour values compared with the control group (P=0.004). At 24 hours, the control group serum values were much lower than those in the moderate-trauma group (P=0.004). In the same group, the moderate-trauma group serum values at 12 hours were significantly higher than those at 0 hours (P=0.002) (Table 2).
In the severe-trauma group, serum UCH-L1 levels were lowest in the control group and highest at 24 hours of trauma (P=0.025). In the pairwise comparisons of serum UCH-L1 levels between groups, the 0-hour serum values in the severe-trauma group were significantly higher than those in the control group (P=0.001). The serum values of the severe-trauma group at 12 hours were significantly higher than those in the control group (P=0.043). Additionally, serum UCH-L1 levels in the control group were significantly lower than the 24-hour values in the severe-trauma group (P=0.026) (Table 2).
DISCUSSION
The origins and mechanisms of GFAP and other biochemical markers remain unknown. Studies investigating the diagnostic significance of GFAP, particularly in severe TBI, indicate that serum GFAP levels are related to the severity of the trauma and mortality [16]. In research conducted by Lumpkins et al. [17] on adults suffering from severe TBI, serum GFAP levels showed a significant decline by the second day. The study also revealed that patients with TBI exhibited much higher serum GFAP levels compared with a control group. As reported by Papa et al. [18], serum GFAP levels were considerably higher in patients with mild to moderate head trauma. In the study, GFAP was present in serum for up to 4 hours after trauma and correlated with indications of injury severity such as Glasgow Coma Scale (GCS) score, brain computed tomography (CT) lesions, and surgical intervention. In our study, serum GFAP levels measured at 0, 12, and 24 hours after mild, moderate, and severe trauma did not differ significantly among the groups, possibly because GFAP levels peaked outside of our specified time range. This is a constraint that restricts the clinical usefulness of the procedure. This constraint requires attention in future research studies, as the specific timeline for GFAP serum levels after trauma remains unclear. Further clinical research and experiments are required to determine the efficacy of this measurement as a diagnostic tool for classifying TBI in clinical practice.
Another recently studied potential biomarker of head trauma is UCH-L1. The reasons for investigating this biomarker include its high degree of neuronal distribution and its role in protein turnover in neurons. Liu et al. [19] reported that UCH-L1 may be a reliable biomarker of TBI within 2 to 6 hours of trauma. High UCH-L1 levels were sustained in the serum for at least 24 hours after TBI. Therefore, UCH-L1 levels can be used to assess the progression of brain damage or recovery.
Reports of high levels of UCH-L1 after TBI suggest a poor neurological prognosis and an increase in mortality [20]. In an experimental study, UCH-L1 increased in concentration in the damaged cortex 48 hours after cortical injury [21]. Kiiski et al. [22] observed an increase in plasma UCH-L1 levels in patients with subarachnoid hemorrhage due to aneurysm, indicating that this elevation persisted for several days after admission, and high UCH-L1 levels have been attributed to secondary damage associated with detrimental neurological outcomes. Majetschak et al. [23] reported that CSF UCH-L1 levels in TBI patients measured over 7 days were associated with a steady decline among survivors, compared with the increase in CSF UCH-L1 levels in deceased patients. The study emphasized that increased UCH-L1 levels may be associated with worse outcomes.
Brophy et al. [24] carried out a study aimed at determining the relationship between UCH-L1 in the biological fluids of TBI patients and their outcomes; CSF and serum samples were collected every 6 hours after severe TBI and were analyzed using ELISA kits. TBI patients who died within 24 hours of the injury had a significant increase in average CSF and serum levels. That study identified a significant correlation between the median concentrations of UCH-L1 in CSF and serum among patients with severe TBI, indicating that these concentrations correspond to clinical outcomes.
Papa et al. [18] detected serum UCH-L1 levels in 295 TBI patients less than 1 hour after injury and found that they increased in accordance with the severity of the injury compared with the control group. In the same study, serum UCH-L1 levels were slightly elevated in TBI patients with a GCS score of 15, and the highest serum UCH-L1 levels were observed in patients with intracranial injuries, particularly those requiring surgical intervention, as detected by CT scan.
Huang et al. [25] monitored 20 male rats over a 24-hour period following moderate TBI. Blood samples were collected 2 days prior to the TBI and at 3, 6, and 24 hours following the injury. A single CSF sample was collected from the cisterna magna 24 hours after TBI. Serum and CSF samples were analyzed for UCH-L1 and GFAP levels. While serum and CSF levels of GFAP were close to zero in the control group, serum GFAP levels at 3 and 6 hours after TBI were significantly higher. UCH-L1 had significantly increased by 3 hours post-TBI. Both GFAP and UCH-L1 in CSF showed a considerable increase compared with controls 24 hours after TBI.
Our study indicated that UCH-L1 levels in serum were not significantly elevated in the mild-trauma group but were significantly elevated in the moderate- and severe-trauma groups (P=0.087 for mild trauma, P=0.001 for moderate trauma, and P=0.025 for severe trauma). The collected data suggest that serum UCH-L1 levels can provide insights into the molecular mechanisms of acute brain injury, predict damage severity, and indicate potentially critical periods following the injury. However, our study found no significant differences in CSF UCH-L1 values among the trauma groups. Our findings, along with a review of the existing research, indicate that UCH-L1 levels increase in biological serum after TBI and correspond with the severity of the injury. Clinical studies and experimental research support the hypothesis that UCH-L1 can act as a reliable biomarker independently or as part of a potential biomarker panel for different kinds of brain injury. To determine the use of UCH-L1 as a clinical diagnostic and prognostic tool, a wide variety of animal models must be studied in greater depth.
The ability to predict the severity of trauma could prevent unnecessary hospital stays and reduce the costs of treatment. Patients with normal test results, such as those produced by a CT scan, but with high serum UCH-L1 levels above the cutoff value can be admitted for monitoring, and neurological damage at the acute phase of trauma can be predicted.
IL-6 is believed to play a crucial role in severe head trauma and in the release of hepatic acute phase proteins such as C-reactive protein, haptoglobin, and ceruloplasmin, while simultaneously reducing albumin synthesis. Several in vitro studies have shown that IL-6 can damage the blood-brain barrier [11]. Although IL-6 is highly sensitive to brain injury, data on its predictive ability are conflicting, particularly in patients with multiple TBIs [12]. In our study, fluctuations in serum IL-6 values were observed in the moderate-trauma group (P=0.019). Fluctuations in IL-6 levels (increasing after trauma, followed by a decrease and then another increase at 24 hours) may be related to a cytokine surge. Jonker et al. [26], in a study examining posttraumatic cytokine levels, reported fluctuations in inflammatory parameters after trauma and emphasized that there could be increasing patterns, particularly 6 to 10 hours later.
Maier et al. [11] found increased IL-6 levels in plasma and CSF after trauma, which remained high for 1 to 3 days and then decreased from day 4 to day 6. During the observations, IL-6 levels were significantly higher in both CSF and plasma in trauma patients than in the control group. However, these CSF levels were greater than those in the plasma in all patients. They concluded that severe TBI triggers an inflammatory response in the brain, in agreement with our findings. In our study, a significant increase was observed in the severe-trauma group at 12 hours (P=0.010), while no significant results were obtained from mild- and moderate-trauma groups.
In a separate study, blood samples from TBI patients at 24 and 48 hours were compared with those of a control group. Patients with mild and moderate head trauma, showed no increase in IL-6 levels during the first 24 hours compared with the control group, but a significant increase was seen at 48 hours [27].
A recent study found that release patterns of IL-6 during the posttraumatic period correlate with severity of trauma. A study featuring 34 patients with TBI individuals grouped according to GCS scores: severe and moderate trauma. The posttraumatic 15-day period was divided into six intervals, during which levels of IL-6 and IL-8 were analyzed. Both IL-6 and IL-8 levels showed a positive correlation with trauma severity. The IL-6 levels peaked on the second day after trauma in the severe-trauma group, while a slower increase was observed in the moderate-trauma group and peaked at day 7. Additionally, early in the posttrauma period, IL-6 levels were more informative than IL-8 levels [28].
In our study, the lack of significant results for serum IL-6 levels may be attributable to the 24-hour limitation of our research. Future studies with a larger sample size and a longer follow-up period could achieve more meaningful results.
In a study conducted by Cikriklar et al. [29], an examination of posttraumatic brain tissue in rats revealed histopathological findings of vascular congestion, hemorrhage, edema, and axonal swelling. Hematoma, axonal structural damage, and edema in our study align with those findings. Cikriklar et al. [29] also reported apoptosis after focal contusion. In their study, the number of apoptotic cells in the cortex increased significantly 2 hours after TBI. The number of apoptotic cells increased with the severity of trauma. It has also been reported that neuronal necrosis occurs, and less-damaged neurons follow the apoptotic pathway.
In a study by Cernak et al. [30], experimentally induced TBI showed neuronal cell death with characteristics of both cell necrosis and apoptosis. Apoptosis and caspase-3 activation were observed in the cerebral hemispheres and brainstem 24 hours after trauma. As a result of our histopathological evaluation, hematoma, neuroinflammation (gliosis), increased vascularity, structural deterioration in neurons, deterioration of axon structure, and edema were more prevalent in the severe-trauma group compared with the other groups. Pathological findings increased from the mild-trauma group to the severe-trauma group in all samples studied. Sample preparations stained with TUNEL revealed that the number of apoptotic cells increased from the mild-trauma group to the severe-trauma group.
Limitations
Our study has some limitations. First, while a rabbit model can provide valuable insights into the pathophysiology of TBI, inherent anatomical and physiological differences between rabbits and humans constrain direct generalizability of our findings to clinical TBI cases. Second, although the power analysis we used had sufficient sample sizing, the number of subjects per group was relatively small, which may limit the effect size of statistical analyses, and this could affect the robustness of the conclusions. Although unexpected fluctuations were observed in the biomarkers in our study, this problem was overcome using nonparametric tests due to the non-normal distribution of the data and by performing intergroup comparisons based on rank scores. Future studies with larger cohorts may validate these results. Finally, certain methodological factors present constraints; the chosen sampling time points of 0, 12, and 24 hours may not have captured the peak expression levels for all biomarkers, as their release kinetics can differ significantly. Furthermore, the inherent variability of ELISA-based assays could have introduced a degree of variance into the biomarker measurements. Acknowledging these limitations helps to place our findings in the proper context and highlights avenues for future research.
Conclusions
In this rabbit head trauma model, serum UCH-L1 levels were higher in both moderate- and severe-trauma groups than in the control group, whereas GFAP and IL-6 levels in CSF rose only after severe trauma. These findings suggest that serum UCH-L1 may help distinguish moderate or greater injury as a biomarker for TBI, while CSF GFAP and IL-6 signal extensive tissue damage. Further studies are needed to confirm these patterns and to assess their clinical relevance.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Supplement1
Print 
