MS

Journal of Proteomics 133 (2016) 93–99

Contents lists available at ScienceDirect

Journal of Proteomics journal homepage: www.elsevier.com/locate/jprot

Quantitative proteomics analysis to identify diffuse axonal injury biomarkers in rats using iTRAQ coupled LC–MS/MS Peng Zhang a, Shisheng Zhu c, Yongguo Li a,b, Minzhu Zhao a,b, Meng Liu a,b, Jun Gao a,b, Shijia Ding b,d, Jianbo Li a,⁎ a

Department of Forensic Medicine, Faculty of Basic Medical Sciences, Chongqing Medical University, Chongqing 400016, China Chongqing Engineering Research Center for Criminal Investigation Technology, Chongqing 400016, China Faculty of Medical Technology, Chongqing Medical and Pharmaceutical College, Chongqing 401331, China d Key Laboratory of Clinical Laboratory Diagnostics (Ministry of Education), College of Laboratory Medicine, Chongqing Medical University, Chongqing 400016, China b c

a r t i c l e

i n f o

Article history: Received 16 July 2015 Received in revised form 17 November 2015 Accepted 17 December 2015 Available online 19 December 2015 Keywords: Diffuse axonal injury Proteomics Rats iTRAQ Biomarkers

a b s t r a c t Diffuse axonal injury (DAI) is fairly common during a traumatic brain injury (TBI) and is associated with high mortality. Making an early diagnosis, appropriate therapeutic decisions, and an accurate prognostic evaluation of patients with DAI still pose difficulties for clinicians. The detailed mechanisms of axonal injury after head trauma have yet to be clearly defined and no reliable biomarkers are available for early DAI diagnosis. Therefore, this study employed an established DAI animal model in conjunction with an isobaric tag for relative and absolute quantification (iTRAQ)-based protein identification/quantification approach. Alterations in rat cerebral protein expression were quantified using iTRAQ coupled LC–MS/MS, with differentially expressed proteins between the control groups, sham and sham-injured, and the injury groups, animals that died immediately post-injury and those sacrificed at 1 h, 6 h, 1 d, 3 d and 7 d post-injury, identified. A total of 1858 proteins were identified and quantified and comparative analysis identified ten candidate proteins that warranted further examination. Of the ten candidate DAI biomarkers, four proteins, citrate synthase (CS), synaptosomal-associated protein 25 (Snap25), microtubule-associated protein 1B (MAP1B) and Rho-associated protein kinase 2 (Rock2), were validated by subsequent Western blot and immunohistochemistry analyses. Our studies not only identified several novel biomarkers that may provide insight into the pathophysiological mechanisms of DAI, but also demonstrated the feasibility of iTRAQ-based quantitative proteomic analysis in cerebral tissue research. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Diffuse axonal injury (DAI), which is predominantly caused by vehicular accidents or other traumas, accounts for up to 70% of all traumatic brain injury (TBI) cases [1] and is associated with high mortality [2]. DAI was initially described as diffuse degeneration of cerebral white matter by Strich, but now the definition of widespread damage of axons in the white matter of the brain is generally accepted [3]. Rotational accelerations are seen as the main contributor to DAI in a traumatic brain injury [2]. Microscopically DAI is characterized by various degrees of derangement of axons and the form of axonal retraction balls. Axonal retraction balls, which are formed in 12–24 h after the injury, are the most striking feature of morphologic changes to axons [4]. It has long been assumed that the direct shear or tensile forces generated by the injury physically disrupt the axons leading to the disconnection of the axons and finally, the form of axonal retraction balls, but recent study showed that the subsequent biochemical cascades postinjury play an important role in the processes of axonal damage [5]. ⁎ Corresponding author at: Chongqing Medical University, Chongqing 400016, China. E-mail address: [email protected] (J. Li).

http://dx.doi.org/10.1016/j.jprot.2015.12.014 1874-3919/© 2015 Elsevier B.V. All rights reserved.

While many studies have examined the pathophysiological mechanisms of DAI, it is still difficult for clinicians to make an early diagnosis, determine appropriate therapeutic interventions, or provide accurate prognostic evaluations. These difficulties are in part attributed to the microscopic and disseminated nature of axons, making detection via imaging very difficult, and the fact that patients show no specific neurological symptoms and can rapidly fall unconscious [6]. Therefore, a need to elucidate the mechanisms leading to axonal injury after a head trauma and identifying reliable biomarkers is necessary to enable early DAI diagnosis. Currently, trauma severity and potential DAI are assessed using magnetic resonance imaging (MRI) and computed tomography (CT). While traditional MRIs are significantly more sensitive than CT scans for detecting TBI associated damage, particularly nonhemorrhagic damage [7–8], histological studies have demonstrated that axonal injuries can still be present with normal MRI results [9–10]. However, histopathological studies cannot identify DAI in patients that die prior to the formation of axonal retraction balls [11]. Therefore, gaining an understanding of the molecular mechanisms driving DAI and identifying differentially expressed proteins associated with axonal injury may facilitate improved DAI diagnoses and clinical management.

94

P. Zhang et al. / Journal of Proteomics 133 (2016) 93–99

Cerebral tissue was chosen in our proteomic analysis work for the possibility that it may elucidate more of the mechanisms of axonal injury in DAI than cerebrospinal fluid or blood. In contrast to genomics and transcriptomics, proteomic analysis can provide insight into the signal transduction events that directly impact the biochemical processes of life. Modern proteomic platforms offer the advantages of high-throughput, simultaneous detection under a single condition and a reduced time and cost. Furthermore, these platforms have already been successfully applied to identify biomarkers associated with many diseases such as colorectal cancer [12], epileptogenesis [13], and oligodendroglioma [14]. Of these approaches, this study utilized tandem mass spectrometry (MS/MS) coupled with isobaric tags for relative and absolute quantification (iTRAQ) labeling to enable the identification and quantification of differentially expressed proteins in a DAI rat model. Bioinformatic analysis was employed to identify differently expressed proteins associated with the mechanisms of axonal injury after head trauma. Overall, these findings will further an understanding of the pathophysiological mechanisms underlying the substantial mortality and specific symptoms in patients with DAI. Furthermore, the identified differentially expressed proteins may serve as reliable DAI biomarkers and aid in early diagnosis and treatment.

the others rapidly progressed into a coma (4.80 ± 1.48 min). The animals that survived the injury were randomly selected and sacrificed at survival periods of 1 h (n = 9), 6 h (n = 9), 1 d (n = 9), 3 d (n = 9) and 7 d (n = 9), which may provide more relevant proteins of DAI and get further investigations into the processes of axonal injury after a head trauma. Nine sham-injured animals underwent the same surgical procedure, including anesthesia and scalp incision, but were not subjected to injury. Nine sham animals received no treatment. Nine of the fifteen rats that died immediately were randomly selected for further study. 2.2.3. Groups Group 1 (n = 18): control group [with two subgroups: sham group (n = 9) and sham-injured group (n = 9)]. Group 2 (n = 54): injury group [with six subgroups: rats that immediately died after injury (n = 9) and those sacrificed at 1 h (n = 9), 6 h (n = 9), 1 d (n = 9), 3 d (n = 9) and 7 d (n = 9) post-injury].

2. Materials and methods

2.2.4. Sample collection Rat cerebral tissues were rapidly removed and three samples from each subgroup were snap frozen in liquid nitrogen and stored at − 80 °C until use. The remaining samples were formalin fixed and used for histopathological analysis and iTRAQ-identified candidate proteins verification.

2.1. Ethics statement

2.3. Histopathological analysis

The procedures for this study were performed according to the Guide for the Care and Use of Laboratory Animals and were approved by the Experimental Animal Care and Use Committee of Chong Qing Medical University.

One of the formalin-fixed cerebral tissue samples in each subgroup was randomly selected for Bielschowsky silver staining.

2.2. Experimental model of diffuse axonal injury

Three frozen cerebral tissues from each of the eight subgroups were homogenized on ice, with 100 g of homogenate from each used for proteomic screening. Protein lysates were obtained using lysis buffer (0.2% Bio-Lyte, 0.001% bromophenol blue, 8 M urea, 4% CHAPS, and 65 mM DTT) followed by centrifugation at 12,000 rpm for 60 min at 4 °C to remove cellular debris. Protein concentrations were established using a Bradford assay, with protein samples (250 mg) digested with trypsin solution (50 μg/ml) overnight at 37 °C and labeled with iTRAQ reagents (8-plex; AB SCIEX, Massachusetts, USA) as follows: sham control (113 tag), sham-injured control (114 tag), rats that died immediately post-injury (115 tag) and rats sacrificed at 1 h (116 tag), 6 h (117 tag), 1 d (118 tag), 3 d (119 tag) and 7 d (121 tag) post-injury according to the manufacturer's instructions. Then, the eight labeled samples were pooled, centrifuged and dried.

2.2.1. Animals A total of 90 adult male Sprague–Dawley rats (350–375 g, housed under 12-h light and dark conditions with food and water available ad libitum) were used in this study. All animals were provided by the C.Q. Medical University Laboratory Animal Center. 2.2.2. Experimental model DAI was induced in animals using an injury model adapted from Marmarou et al. [15–16]. A height of 150 cm was chosen in this injury model instead of 200 cm adopted by Marmarou et al. for the reduction in injury intensity and mortality to obtain less severe DAI which often are undiagnosed [6]. For this procedure, 72 randomly selected adult male Sprague–Dawley rats were anesthetized via intraperitoneal injections of chloral hydrate (10%, 0.35 ml/100 g). After achieving the appropriate level of anesthesia, the animal's scalp was shaved, and a midline incision was performed to expose the vertex of the skull. A 10-mm-diameter stainless steel disk with a thickness of 3 mm was affixed to the skull at the midline between the bregma and lambdoid sutures using dental acrylic. The incision was sutured, and the animals were housed normally. The injury device was prepared the next day and consisted of a vertical plexiglas tube with a 25 mm outer diameter that was attached to a ring stand. After rats were lightly anesthetized with ether, they were placed on a foam pad in a prone position. The hammer (weighing 450 g) was allowed to fall freely from a height of 150 cm directly onto the stainless steel disk fixed to the animal's skull. The foam pad was quickly removed immediately after impact to avoid a second impact and the metallic disk was removed. The animals' skulls were examined for any sign of fracture, with animals exhibiting fractures eliminated from further study. The incisions were sutured and respiration and heart rates were monitored and recorded every 5 min. Fifteen rats died post-injury (20.8% mortality rate) due to central respiratory depression as described by Marmarou et al. [15–16], while

2.4. iTRAQ sample preparation

2.5. LC and MS/MS analyses Nano-LC–MS/MS experiments were performed on a Triple-TOF 5600 system (AB SCIEX, Massachusetts, USA) with iTRAQ-labeled peptide mixtures separated using an Eksigent nanoLC Ultra binary pump system with tray cooling. Peptides were loaded on a 75 μm × 150 mm nanocolumn and a 200 μm × 0.5 mm nanotrap column (Eksigent), both filled with ChromXP C18-CL 3 μm 120 Å phase, followed by a mobile phase elution with buffer A (2% ACN and 0.1% FA) and buffer B (98% ACN and 0.1% FA). Peptides were then eluted in a linear gradient with buffer B from 12% to 32% over 90 min at a flow rate of 300 nl/min. MS analysis was performed on the liquid chromatographic eluent at a mass spectra range of 350–1250 m/z and an accumulation time of 250 ms per spectrum. 2.6. Proteomic data analysis and bioinformatics The Analyst QS 1.1 software (Applied Biosystems) was used to acquire the iTRAQ-based proteomics data. Peptides were identified

P. Zhang et al. / Journal of Proteomics 133 (2016) 93–99

and quantified using ABI ProteinPilot software 4.0 (AB SCIEX). The Paragon database search algorithm [17] and an integrated false discovery rate (FDR) analysis [18] were implemented in the ProteinPilot software for peptide identification. The user defined parameters included (i) sample type, iTRAQ 8-plex (peptide labeled); (ii) cysteine alkylation, iodoacetamide; (iii) digestion, trypsin; (iv) instrument, Triple-TOFTM 5600; (v) special factors, none; (vi) species, none; (vii) ID focus, biological modifications, amino acid substitutions; (viii) database, RAT.4.5.0.0,1654; and (ix) search effort, thorough. A 95% confidence interval (CI) was set as the significance threshold for protein identification. The protein lists from the 8 iTRAQ experiments were merged with ratios calculated based on the sham control-113. The confidence level of each differentially expressed protein was calculated as a p-value using ProteinPilot, allowing the results to be evaluated based not only on the magnitude of the change but also on the confidence level of the change. Gene ontology (GO) was performed using the bioinformatics analysis tool DAVID (http://david.abcc.ncifcrf.gov) to determine the functional classifications of the iTRAQ-identified proteins.

95

2.8. Immunohistochemistry Immunohistochemistry (IHC) was performed on the remaining 40 formalin-fixed cerebral tissue samples to validate the four candidate biomarkers. The primary antibodies included polyclonal anti-rat antibodies to Snap-25, Rock2, citrate synthase (CS) and microtubuleassociated protein 1B (MAP1B) (all at 1:200; San Ying Biotechnology, Inc., Wuhan, China). Five regions within each IHC stained tissue section were randomly selected by two independent investigators for evaluation. Tissue section was scored based on the degree of immunopositive staining as follows: b10% (level 0), 10–30% (level 1), 30–50% (level 2), 50–70% (level 3) and N70% (level 4) and staining intensities were categorized into four levels: none (level 0), mild (level 1), moderate (level 2) and intense (level 3) [14]. These assigned sample scores were used for quantitative analysis and a Mann–Whitney–Wilcoxon U test was used for statistical analysis, with a two-sided p-value b0.05 considered statistically significant.

3. Results 2.7. Western blotting validation Western blot analyses were performed to validate the iTRAQ-based results [19]. Proteins from each subgroup (30 μg) were separated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis for 2 h. The proteins were then transferred to a PVDF membrane by semi-dry blotting (Bio-Rad) at 90 V with carbonate transfer buffer. The membranes were blocked using evaporated milk for 1 h and then incubated overnight at 4 °C with primary polyclonal antibodies, including antibodies against synaptosomal-associated protein 25 (Snap-25), Rho-associated protein kinase 2 (Rock2) and β-actin (both at 1:1000; Bioworld Technology, Inc., Louis Park, USA). The membranes were then incubated for 1 h at room temperature with secondary antibodies (1:5000; Bioworld Technology, Inc., Louis Park, USA). An ECL detection system (Amersham Biosciences) was used for visualization and protein quantification was performed using an ImageQuant image analysis system (Storm Optical Scanner, Molecular Dynamics), three times was repeated and the data was normalized to the average.

3.1. Identification of differentially expressed proteins via iTRAQ coupled LC–MS/MS DAI animal models were first histopathological validated (Fig. 1) and three cerebral tissue samples from each subgroup were randomly selected for iTRAQ coupled MS/MS as a means to identify candidate proteins associated with the pathophysiological mechanisms of axonal injury. A total of 1858 proteins were identified and quantified overall, with GO analysis groupings including biological processes (BP), cellular component association (CC), and molecular function (MF; Fig. 2A–C). Proteins with expression ratios of over 1.2-fold in increase or at least 0.8-fold in decrease while p-value b0.05 were considered differentially expressed. Following comparative analysis, ten differentially expressed proteins were identified between the control and injury groups (p-value b 0.05; Table 1). A representative MS/MS spectra and the relative intensity of the reporter ions of a differentially expressed protein are shown in Fig. 3 (Rock2, up-regulated in DAI cerebral tissue).

Fig. 1. Histopathological analysis of cerebral tissue samples from the control and injury groups. (A) sham group; (B) sham-injured group; (C) immediately died after injury; and (D–G) injured and sacrificed at 1 h, 6 h, 1 d, 3 d and 7 d post-injury. All the samples in each group were obtained from corpus collosum. Tissue examination was performed following Bielschowsky silver staining. In the injury group, the axons showed various degrees of derangement, intermittent swellings and axonal retraction balls as early as 6 h post-injury (D–G), with normal axons noted in the controls (A–B).

96

P. Zhang et al. / Journal of Proteomics 133 (2016) 93–99

Fig. 2. Gene ontology analysis for the 1858 proteins identified and quantified by iTRAQ. (A) Genological functions of the proteins; (B) cellular components of the proteins; and (C) molecular functions of the proteins.

3.2. Western blot validation of iTRAQ-based results iTRAQ coupled MS/MS results were validated by examining Rock2 and Snap-25 expression via Western blot analysis. While the foldchange results from these two methods varied slightly, the Western blot analysis also showed an up-regulation of Rock2 and a downregulation of Snap-25 in DAI cerebral tissue relative to the control group (Fig. 4).

3.3. Validation of candidate biomarkers A subset of the identified candidate biomarkers, including CS, Snap25, Map1B and Rock2, was selected for IHC analysis based on their possible functions in axonal injury during DAI. The remaining five

formalin-fixed cerebral tissue samples in each subgroup were analyzed. Relative to the control group, the injury group showed significantly lower levels of Snap-25 and CS staining and higher levels of Map1B and Rock2 staining (Fig. 5). 4. Discussion TBI biomarkers from proteomic studies can provide clinicians with a powerful tool for early diagnosis, early assessment of severity, and accurate prognostic evaluation of patients with brain injury [20–24], especially for mild TBI (mTBI) which often shows no visible focal lesions detected by routine imaging techniques [25]. To date, some proteins have been identified as candidate TBI biomarkers, such as ubiquitin carboxy-terminal hydrolase L1 protein (UCHL1) [26], neuron-specific enolase (NSE) [27], and S100-B [28]. DAI is a special type of TBI which

P. Zhang et al. / Journal of Proteomics 133 (2016) 93–99

97

Table 1 Differentially expressed proteins in the cerebral tissue of the injury group identified using an iTRAQ-based quantitative proteomics approach. No.

Gene name

Uniprot ID

Time point

Fold change

p-Value

Down-regulated in cerebral tissue with DAI 1 Synaptosomal-associated protein 25 2 Mitochondrial glutamate carrier 2 3 Syntaphilin 4 Plectin 5 Citrate synthase 6 Aconitate hydratase 7 Myosin-10 8 Neurochondrin

Protein name

Snap-25 GHC2 Snph Plec Cs Aco2 Myh10 Ncdn

P60881 Q505J6 B5DF41 P30427 Q8VHF5 Q9ER34 Q9JLT0 O35095

3d 7d 1d 1d 1h 1h 1d 1d

0.46 0.57 0.14 0.54 0.69 0.76 0.50 0.55

0.045183849 0.033457011 0.007920128 0.045268271 0.03464213 0.031120639 0.036413319 0.031622712

Up-regulated in cerebral tissue with DAI 9 Rho-associated protein kinase 2 10 Microtubule-associated protein 1B

Rock2 MAP1B

Q62868 P15205

7d 3d

1.62 1.20

0.047817718 0.042255301

is characterized by widespread damage of axons which often are undiagnosed [6]. The primary objectives of this study were to examine DAI pathophysiological mechanisms and identify candidate biomarkers to improve early clinical diagnoses and therapies. While previous studies have searched for sensitive DAI biomarkers, none have been identified. In this study, iTRAQ coupled MS/MS was utilized to examine rat cerebral tissues to identify specific candidate DAI biomarkers. A comparative protein analysis between control and injury groups identified ten candidate proteins that warranted further examination. Among these, four candidate biomarkers were selected based on their biological functions during axonal injury progression and their high fold-changes. All four of these candidate biomarkers were successfully validated using Western blot analysis and IHC and may significantly contribute to a further understanding of the complex pathophysiological mechanisms of DAI. Although we cannot ensure that the four candidate biomarkers selected were not associated with other processes, such as increase of intracranial pressures, decrease of cerebral blood flow, edema, cellular swelling and other pathophysiological changes that occur in DAI, their critical functions in the mechanisms of axonal injury showed a great possibility that the four differentially expressed proteins can be used as candidate biomarkers of DAI in rats. This study utilized iTRAQ which offers the ability to simultaneously detect eight samples under a single condition [29]. Moreover, isobaric labeling for longitudinal (time-course) studies have great strengths in mitigating cell- and cell-cycle-specific protein expression patterns which might confound proteomics studies that are not longitudinal [29]. This approach enabled the identification of 1858 proteins, with

10 of those proteins found to be differentially expressed between the control and injured groups. Of the identified differentially expressed proteins, 4 were further validated via Western blot and IHC. One of the candidate biomarkers examined was citrate synthase (CS), which was significantly down-regulated 1 h post-injury in DAI cerebral tissue relative to the sham animals. CS functions by binding to the inner membrane surface of the mitochondria and plays a critical role in the central metabolic pathway of aerobic organisms, the tricarboxylic acid (TCA) cycle. The down-regulation of CS that was noted herein may be associated with a decrease in cerebral blood flow and mitochondrial dysfunction, thus suggests an early reduction in energy supplies post-injury. However, previous studies have shown that TBI can cause an increased energy demand due to a disturbance of ion homeostasis [30–31], glutamatergic neurotransmission [32] and the activation of energyconsuming repair processes [33]. Axonal injury can be divided into primary and secondary axotomies which are caused not only by the direct mechanical insult, but also by a progressive molecular and cellular cascade of pathological changes within the axon after the initial shear stresses at the time of injury [5]. Therefore, an imbalance between energy supply and demand may be a major early contributor to the secondary injury cascade, which plays a pivotal role in the outcome of DAI patients. Moreover, cerebral energy metabolism impairment, which could occur soon after injury, also indicates that increasing the energy supply may help reduce axonal injury and improve patients' clinical prognoses. Another identified candidate biomarker was Snap-25, which was significantly down-regulated 3 d post-injury. Snap-25 can be found in

Fig. 3. Representative differentially expressed protein peptide spectra. (A) MS/MS spectrum of a Rock2 peptide (QIFGDYK) that was up-regulated in DAI cerebral tissue. (B) Spectrum showing the relative reporter ion intensity for the above peptide. [m/z; sham control (113 tag); sham-injured sample (114 tag); animals that died immediately post-injury (115 tag) and those sacrificed at 1 h (116 tag), 6 h (117 tag), 1 d (118 tag), 3 d (119 tag) and 7 d post-injury (121 tag)].

98

P. Zhang et al. / Journal of Proteomics 133 (2016) 93–99

Fig. 4. Western blot analysis to verify selected differentially expressed proteins. Candidate proteins were examined in triplicate and normalized to β-Actin levels (loading control) for quantitative analysis.

the axonal plasmalemma and the intracellular vesicles of neurons [34–35]. This protein not only plays an important role in the regulation of synaptic vesicle exocytosis [36–38], but is also involved in axonal elongation [39]. Furthermore, recent studies have shown that Snap-25 may be associated with various psychiatric and neurological disorders such as schizophrenia, attention-deficit hyperactivity disorder (ADHD), and epilepsy [40–41]. In the present study, Snap-25 levels decreased between 1 h and 6 h post-injury, reaching a 70% reduction relative to the control at 6 h. From 6 h to 1 d, Snap-25 did not decline significantly, until it then reached its lowest level (46% of the control level) at 3 d post-injury. This decreased Snap-25 expression over time

implies that while the primary axotomy is associated with the initial injury, the secondary axotomy most likely occurs between 6 h and 1 d post-injury, thus providing a substantial therapeutic window for intervention to impede neurodegeneration or encourage regeneration. In the present study, MAP1B was the first significantly up-regulated protein post-injury. MAP1B aids in microtubule-stabilization and is highly expressed in developing neurons and down-regulated in the adult nervous system [42]. Recent studies have also shown that MAP1B is up-regulated after ischemia in aged rats, which may indicate an attempt to repair or regenerate the tissue [43]. During DAI, the upregulation of MAP1B may also be attributed to post-injury ischemia of the cerebral tissue and may play an important role in maintaining axon stability. These findings suggest that promoting the expression of MAP1B after an axonal injury may reduce axotomy and improve clinical outcome. Rock2 is predominantly expressed in the brain and heart [44–45] and was found to be significantly up-regulated in cerebral DAI tissues in this study. Rock is a downstream effector of the GTPase Rho and ultimately mediates actin remodeling within the cytoskeleton in response to extracellular signals. Additionally, Rock2 regulates the myosin light chain (MLC) via direct phosphorylation and inactivates myosin phosphatase through the phosphorylation of the myosin binding subunit (MBS) [46]. In the brain, Rock2 appears to be a key link between actin filament dynamics and axonal growth inhibition [47]. Through the phosphorylation of downstream targets such as intermediate filament proteins, microtubule-associated proteins (MAPs), and collapsin response mediator protein 2 (CRMP2), active Rock2 regulates the depolymerization of actin and induces neurite retraction and growth cone collapse, which ultimately inhibits axonal growth. The noted Rock2 up-regulation in cerebral tissue post-injury may be associated with the activation of Rho family GTPases during ischemia and hypoxia. Additionally, the activation of the Rho signaling pathway could induce vasospasm and then aggravate ischemia and hypoxia in the cerebral tissue. Recent studies have shown that Rho signaling limits axonal growth and neurological recovery after a central nervous system (CNS) injury in adult mammals, while the suppression of Rho-kinase activity may enhance axonal regeneration in the CNS [48]. A number of axonal injury marker candidates have been reported in previous studies following TBI, such as amyloid-β (Aβ) peptides [24,49], tau protein [24,49] and neurofilament light polypeptide [49]. Although significant changes were observed in patients and/or animal models in previous TBI studies, biomarkers of axonal injury following TBI were

Fig. 5. Immunohistochemical validation of four candidate biomarkers. Snap-25 (A–B), CS (C), MAP1B (D) and Rock2 (E) were examined using the remaining 40 formalin-fixed and paraffin-embedded samples. (A) Immunohistochemistry results for the Snap-25 in the cerebral tissue of the sham, sham-injured and 3 d groups. (B–E) Quantitative scoring results of IHC analyses are shown as box plots.

P. Zhang et al. / Journal of Proteomics 133 (2016) 93–99

not identified as candidate DAI biomarkers in our proteomics-based research. This may be attributed to the difference of injury models, injury severity and the time span post-injury chosen in the studies. In conclusion, this study utilized an iTRAQ-based proteomic approach as a means to identify several differentially expressed proteins in DAI cerebral tissues. Four candidate DAI biomarkers, CS, Snap25, MAP1B and Rock2, were further validated via Western blot and IHC analyses. However, future studies are necessary to determine whether these findings can be extended to human patients. Therefore, we will perform additional studies with these proteins to further examine the pathophysiological mechanisms of DAI. Conflict of interest The authors have declared that no competing interests exist. Acknowledgments This work was funded by the National Natural Science Foundation of China (81273344), the Open Project of Shanghai Key Laboratory of Forensic Medicine (KF1105) and Application Development Plan Project of Chongqing (cstc2014yykfA110003).

[20] [21]

[22]

[23]

[24]

[25] [26]

[27] [28]

[29]

References [30] [1] G.S. McHugh, D.C. Engel, I. Butcher, E.W. Steyerberg, J. Lu, N. Mushkudiani, et al., Prognostic value of secondary insults in traumatic brain injury: results from the IMPACT study, J. Neurotrauma 24 (2007) 287–293. [2] X.Y. Li, D.F. Feng, Diffuse axonal injury: novel insights into detection and treatment, J. Clin. Neurosci. 16 (2009) 614–619. [3] J.H. Adams, D.U. Graham, L.S. Murray, G. Scott, Diffuse axonal injury due to nonmissile head injury in humans: an analysis of 45 cases, Ann. Neurol. 12 (1982) 557–563. [4] S. Kallakuri, J.M. Cavanaugh, A.C. zaktay, T. Takebayashi, The effect of varying impact energy on diffuse axonal injury in the rat brain: a preliminary study, Exp. Brain Res. 148 (2003) 419–424. [5] G.A. Dienel, Lactate shuttling and lactate use as fuel after traumatic brain injury: metabolic considerations, J. Cereb. Blood Flow Metab. 34 (2014) 1736–1748. [6] P.M. Kochanek, R.P. Berger, H. Bayir, A.K. Wagner, L.W. Jenkins, R.S. Clark, Biomarkers of primary and evolving damage in traumatic and ischemic brain injury: diagnosis, prognosis, probing mechanisms, and therapeutic decision making, Curr. Opin. Crit. Care 14 (2008) 135–141. [7] A. Mallouhi, Craniocerebral trauma: magnetic resonance imaging of diffuse axonal injury, J. Radiol. 54 (2014) 907–915. [8] M. Smits, M.G. Hunink, D.A. van Rijssel, Outcome after complicated minor head injury, Am. J. Neuroradiol. 29 (2008) 506–513. [9] M.E. Shenton, H.M. Hamoda, J.S. Schneiderman, S. Bouix, O. Pasternak, Y. Rathi, et al., A review of magnetic resonance imaging and diffusion tensor imaging findings in mild traumatic brain injury, Brain Imaging Behav. 6 (2012) 137–192. [10] J. Li, X.Y. Li, D.F. Feng, L. Gu, Quantitative evaluation of microscopic injury with diffusion tensor imaging in a rat model of diffuse axonal injury, Eur. J. Neurosci. 33 (2011) 933–945. [11] A. Bisht, K. Garg, D. Agarwal, P.K. Singh, G.D. Satyarthee, D. Gupta, et al., Histological changes in thalamus in short term survivors following traumatic brain injury: an autopsy study, Neurol. India 61 (2013) 599–605. [12] P. Alvarez-Chaver, O. Otero-Estévez, M. Páez de la Cadena, F. Rodríguez-Berrocal, V.S. Martínez-Zorzano, Proteomics for discovery of candidate colorectal cancer biomarkers, World J. Gastroenterol. 20 (2014) 3804–3824. [13] K. Lukasiuk, A.J. Becker, Molecular biomarkers of epileptogenesis, Neurotherapeutics 11 (2014) 319–323. [14] L.S. Yang, X. Xu, X.P. Liu, H. Jin, Z.Q. Chen, X.H. Liu, et al., iTRAQ-based quantitative proteomic analysis for identification of oligodendroglioma biomarkers related with loss of heterozygosity on chromosomal arm 1p, J. Proteome 77 (2012) 480–491. [15] A. Marmarou, M.A. Foda, W. van den Brink, J. Campbell, H. Kita, K. Demetriadou, A new model of diffuse brain injury in rats. Part I: pathophysiology and biomechanics, J. Neurosurg. 80 (1994) 291–300. [16] M.A. Foda, A. Marmarou, A new model of diffuse brain injury in rats. Part II: morphological characterization, J. Neurosurg. 80 (1994) 301–313. [17] I.V. Shilov, S.L. Seymour, A.A. Patel, A. Loboda, W.H. Tang, S.P. Keating, et al., The Paragon Algorithm, a next generation search engine that uses sequence temperature values and feature probabilities to identify peptides from tandem mass spectra, Mol. Cell. Proteomics 6 (2007) 1638–1655. [18] H. Lam, E.W. Deutsch, R. Aebersold, Artificial decoy spectral libraries for false discovery rate estimation in spectral library searching in proteomics, J. Proteome Res. 9 (2010) 605–610. [19] K.E. Williams, O. Miroshnychenko, E.B. Johansen, R.K. Niles, K.K. SundaramR, et al., Urine, peritoneal fluid and omental fat proteomes of reproductive age women:

[31]

[32]

[33] [34] [35]

[36]

[37]

[38] [39] [40]

[41]

[42] [43] [44]

[45] [46]

[47]

[48]

[49]

99

endometriosis-related changes and associations with endocrine disrupting chemicals, J. Proteome 15 (2015) 194–205. S. Shen, R.R. Loo, I.B. Wanner, J.A. Loo, Addressing the needs of traumatic brain injury with clinical proteomics, Clin. Proteomics 11 (2014) 11. F. Crawford, G. Crynen, J. Reed, B. Mouzon, A. Bishop, B. Katz, Identification of plasma biomarkers of TBI outcome using proteomic approaches in an APOE mouse model, J. Neurotrauma 29 (2012) 246–260. J. Hanrieder, M. Wetterhall, P. Enblad, L. Hillered, J. Bergquist, Temporally resolved differential proteomic analysis of human ventricular CSF for monitoring traumatic brain injury biomarker candidates, J. Neurosci. Methods 177 (2009) 469–478. G. Hergenroeder, J.B. Redell, A.N. Moore, W.P. Dubinsky, R.T. Funk, J. Crommett, et al., Identification of serum biomarkers in brain-injured adults: potential for predicting elevated intracranial pressure, J. Neurotrauma 25 (2008) 79–93. T.M. Evans, H. Van Remmen, A. Purkar, S. Mahesula, J.A. Gelfond, M. Sabia, et al., Microwave & magnetic (M2) proteomics of a mouse model of mild traumatic brain injury, Transl. Proteom. 3 (2014) 10–21. H. Zetterberg, D.H. Smith, K. Blennow, Biomarkers of mild traumatic brain injury in cerebrospinal fluid and blood, Nat. Rev. Neurol. 9 (2013) 201–210. A.M. Boutté, C. Yao, F. Kobeissy, X.C. May Lu, Z. Zhang, K.K. Wang, et al., Proteomic analysis and brain-specific systems biology in a rodent model of penetrating ballistic-like brain injury, Electrophoresis 33 (2012) 3693–3704. S. Mondello, U. Muller, A. Jeromin, J. Streeter, R.L. Hayes, K.K. Wang, Blood-based diagnostics of traumatic brain injuries, Expert. Rev. Mol. Diagn. 11 (2011) 65–78. A. Rodríguez-Rodríguez, J.J. Egea-Guerrero, A. León-Justel, E. Gordillo-Escobar, J. Revuelto-Rey, A. Vilches-Arenas, et al., Role of S100B protein in urine and serum as an early predictor of mortality after severe traumatic brain injury in adults, Clin. Chim. Acta 414 (2012) 228–233. E. Takahashi, A. Okumura, H. Unoki-Kubota, H. Hirano, M. Kasuga, Y. Kaburagi, Differential proteome analysis of serum proteins associated with the development of type 2 diabetes mellitus in the KK-A(y) mouse model using the iTRAQ technique, J. Proteome 84 (2013) 40–51. M. Reinert, B. Hoelper, E. Doppenberg, A. Zauner, R. Bullock, Substrate delivery and ionic balance disturbance after severe human head injury, Acta Neurochir. Suppl. 76 (2000) 439–444. C.L. Osteen, A.H. Moore, M.L. Prins, D.A. Hovda, Age-dependency of 45calcium accumulation following lateral fluid percussion: acute and delayed patterns, J. Neurotrauma 18 (2001) 141–162. N.S. Rozas, J.B. Redell, J.L. Hill, J. McKenna 3rd, A.N. Moore, M.J. Gambello, et al., Genetic activation of mTORC1 signaling worsens neurocognitive outcome after traumatic brain injury, J. Neurotrauma 32 (2015) 149–158. D.H. Smith, D.F. Meaney, W.H. Shull, Diffuse axonal injury in head trauma, J. Head Trauma Rehabil. 18 (2003) 307–316. W.L. Maxwell, J.T. Povlishock, D.L. Graham, A mechanistic analysis of nondisruptive axonal injury: a review, J. Neurotrauma 14 (1997) 419–440. M. Barrenschee, M. Böttner, J. Harde, C. Lange, F. Cossais, M. Ebsen, et al., SNAP-25 is abundantly expressed in enteric neuronal networks and upregulated by the neurotrophic factor GDNF, Histochem. Cell Biol. 143 (2015) 611–623. A. Balkarli, C. Sengül, E. Tepeli, H. Balkarli, V. Cobankara, Synaptosomal-associated protein 25 (Snap-25) gene polymorphism frequency in fibromyalgia syndrome and relationship with clinical symptoms, BMC Musculoskelet. Disord. 15 (2014) 191. R. Tomasoni, D. Repetto, R. Morini, C. Elia, F. Gardoni, M. Di Luca, et al., SNAP-25 regulates spine formation through postsynaptic binding to p140Cap, Nat. Commun. 4 (2013) 2136. S. Yamamori, M. Itakura, Differential expression and function of SNAP-25 family proteins in the mouse brain, Seikagaku 85 (2013) 1016–1020. E.K. Buschbeck, R.R. Hoy, The development of a long, coiled, optic nerve in the stalkeyed fly Cyrtodiopsis whitei, Cell Tissue Res. 321 (2005) 491–504. T.K. Choi, H.S. Lee, J.W. Kim, T.W. Park, D.H. Song, K.W. Yook, et al., Support for the MnlI polymorphism of SNAP25: a Korean ADHD case–control study, Mol. Psychiatry 12 (2007) 224–226. M. Kataoka, S. Yamamori, E. Suzuki, S. Watanabe, T. Sato, H. Miyaoka, et al., A single amino acid mutation in SNAP-25 induces anxiety-related behavior in mouse, PLoS One 6 (2011) e25158. S. Halpain, L. Dehmelt, The MAP1 family of microtubule-associated proteins, Genome Biol. 7 (2006) 224. D. Villarroel-Campos, C. Gonzalez-Billault, The MAP1B case: an old MAP that is new again, Dev. Neurobiol. 74 (2014) 953–971. X.Q. Chen, I. Tan, C.H. Ng, C. Hall, L. Lim, T. Leung, Characterization of RhoA-binding kinase ROKalpha implication of the pleckstrin homology domain in ROKalpha function using regionspecific antibodies, J. Biol. Chem. 277 (2002) 12680–12688. L. Julian, M.F. Olson, Rho-associated coiled-coil containing kinases (ROCK): structure, regulation, and functions, Small GTPases 5 (2014) e29846. R. Kalaji, A.P. Wheeler, J.C. Erasmus, S.Y. Lee, R.G. Endres, L.P. Cramer, ROCK1 and ROCK2 regulate epithelial polarisation and geometric cell shape, Biol. Cell. 104 (2012) 435–451. J.C. Koch, L. Tönges, E. Barski, U. Michel, M. Bähr, P. Lingor, ROCK2 is a major regulator of axonal degeneration, neuronal death and axonal regeneration in the CNS, Cell Death Dis. 5 (2014) e1225. D. Philip, S. Andre, S. Antonio, et al., Rho-associated kinase II (ROCKII) limits axonal growth after trauma within the adult mouse spinal cord, J. Neurosci. 29 (2009) 15266–15276. J. Li, Y.L. Xue, F.F. Dong, C.P. Dong, Biomarkers associated with diffuse traumatic axonal injury: exploring pathogenesis, early diagnosis, and prognosis, J. Trauma 69 (2010) 1610–1618.