Secondary Logo

Journal Logo

Basic Science Aspects

Modulation of Brain Transcriptome by Combined Histone Deacetylase Inhibition and Plasma Treatment Following Traumatic Brain Injury and Hemorrhagic Shock

Dekker, Simone E.∗,†; Biesterveld, Ben E.; Bambakidis, Ted; Williams, Aaron M.; Tagett, Rebecca; Johnson, Craig N.; Sillesen, Martin§,||; Liu, Baoling; Li, Yongqing; Alam, Hasan B.

Author Information
doi: 10.1097/SHK.0000000000001605

Abstract

INTRODUCTION

Despite increased research focus on injuries, traumatic brain injury (TBI) remains a leading cause of morbidity and mortality (1). TBI frequently presents with hemorrhagic shock (HS), creating a combined insult that is especially lethal (2). Although the primary brain injury occurs in the prehospital setting and results in rapid damage to the neuronal tissue, the secondary injury, which evolves over the next few hours to days, can be attenuated with appropriate treatments. On a subcellular level, such secondary injury is thought to result from a major disruption of genome, proteome, and inflammasome homeostasis (3).

Effective treatment strategies should aim to restore these disrupted systems. Yet, crystalloid resuscitation remains in widespread use in trauma care, although it possesses no inherent pro-survival properties and may actually exacerbate hemostatic derangements, endothelial dysfunction, inflammation, and cerebral edema (4, 5). Because of this, early use of blood products is becoming a cornerstone of contemporary trauma resuscitation. Specifically, there have been efforts to use plasma products early in the prehospital settings (6). Clinical studies suggest improved hemostasis and fewer deaths due to exsanguination when a higher ratio of plasma to red blood cells is utilized in early resuscitation (7). In fact, administration of fresh frozen plasma (FFP) has been shown to attenuate neurologic impairment and preserve cognitive functions in preclinical studies (8). While FFP represents an improvement over traditional crystalloid resuscitation, recent studies have demonstrated the addition of histone deacetylase inhibitor therapy yields superior clinical outcomes. Valproic acid (VPA) is a histone deacetylase (HDAC) inhibitor that has been shown to reduce brain swelling and lesion size, and improve long-term outcomes in large animal models of TBI+HS (9, 10).

However, the mechanisms underlying this improved outcome are not fully understood. As VPA is an HDAC inhibitor, the key mechanisms behind this protective effect may be related to gene expression. However, the effect of combined treatment with FFP+VPA on the brain transcriptome is unknown. Moreover, there is evidence that FFP alone acts at the level of gene transcription, affecting the expression of genes involved in inflammation, platelet signaling, and metabolism (11). Emerging evidence suggests that VPA may work synergistically with FFP to further potentiate this neuroprotective effect (12). However, little is known about the type, function, and interaction of the genes and pathways affected by combined FFP+VPA treatment compared with FFP alone. This study uses a systems biology approach to investigate the effects of FFP+VPA resuscitation on the brain transcriptome in a swine model of TBI+HS. We hypothesize that FFP+VPA treatment upregulates pathways associated with histone acetylation, stimulates pro-survival pathways, and attenuates inflammation.

MATERIALS AND METHODS

These experiments were conducted in compliance with the Animal Welfare Act and other federal regulations. The Institutional Animal Care and Use Committee approved the study protocol, and investigators adhered to the standards set forth in the Guide for the Care and Use of Laboratory Animals, Institute for Laboratory Animal Research (1996). A staff veterinarian supervised all procedures and animal care.

Animal model

A previously validated porcine model of TBI+HS was used to compare FFP to FFP+VPA treatment (13). Female Yorkshire swine (42 kg–50 kg, Tufts Veterinary School, Grafton, Mass) were intubated, sedated, and invasively monitored for the entire experiment. A 12 mm depth controlled cortical impact was inflicted on exposed dura as previously described (14). Concurrent to inflicting the TBI, the animal was hemorrhaged 40% of its blood volume over 20 min. Following hemorrhage, the animals were kept in shock (mean arterial pressure of 30 mm Hg–35 mm Hg) for 2 h (simulating delayed medic response in the field), and then resuscitated with FFP or FFP+VPA (n = 5/group). In the FFP+VPA group, VPA was given intravenously 1 h after hemorrhage at 100 mg/kg per hour. Two hours after hemorrhage, each animal was resuscitated with FFP that matched its hemorrhage volume. After treatment, the animals were observed for 6 h and then euthanized.

Tissue collection and ribonucleic acid (RNA) extraction

The brain was harvested as previously described (9). The brain was sliced into 5-mm-thick coronal sections. The coronal section containing the lesion was separated, and each hemisphere was cut into eight radially symmetric pieces emanating from the midline. The piece of tissue immediately adjacent/inferior to the primary lesion was separated and flash-frozen for subsequent RNA extraction. Brain RNA was isolated using the RNeasy Lipid Tissue Mini Kit using the manufacturer's instructions (Qiagen, Hilden, Germany). 30 mg of brain tissue was homogenized and lysed using 1 mL QIAzol Lysis Reagent. After incubation, extraction, and centrifugation steps, the sample was added to the RNeasy Mini spin column. On-column DNase digestion was then performed, and final washed samples were extracted and stored at −80 C.

Microarray

Gene expression was analyzed using an Affymetrix Porcine ST 1.1 microarray (Affymetrix, Santa Clara, Calif) performed by the University of Michigan DNA Sequencing Core Facility (Ann Arbor, Mich), as previously described (15, 16). RNA was prepared for microarray expression analysis using the GeneChip 3’ IVT Expression Kit (Affymetrix Inc, Santa Clara, Calif; 394,580 probes mapping to 19,212 probe sets; median of 22 probes per probe set). Poly-A RNA Control and T7-Oligo(dT) primer was added to the total RNA sample to synthesize first-strand complementary DNA. After second-strand complementary DNA synthesis, biotinylated ribonucleotide analog was used to create artificial RNA. After purification and fragmentation of the biotin-labeled artificial RNA, the sample was hybridized onto the GeneChip 3’ expression array.

Gene ontology analysis

Differentially expressed (DE) genes were identified using a threshold < 0.05 for statistical significance (P value) and a log-fold change in expression > 0.585 (fold-change > 1.5). P values were adjusted for multiple comparisons using false discovery rate (FDR) (17). iPathwayGuide software (Advaita Bioinformatics, Plymouth, Mich) was used to identify significantly impacted Gene Ontology (GO) terms. GO terms were used to describe the highest ranked Biological Processes. Data were analyzed in the context of gene ontologies from the Gene Ontology Consortium database (18). For each GO term, the number of DE genes annotated to the term was compared with the number of DE genes expected by chance. Elim pruning P value correction was used to iteratively eliminate the genes mapped to a significant GO term from more general GO terms (19). A threshold of a minimum of five DE genes per GO term was used. The remaining GO terms were then clustered using a custom method developed by the University of Michigan Bioinformatics Core Facility. GO terms were grouped based on their constituent genes by recursively merging GO terms with fewer DE genes (Tsm) into GO terms with more DE genes (Tlg). We set two strict criteria to merge Tsm into Tlg: Tsm must share ≥ 95% of its constituent genes with Tlg; the number of differentially expressed genes in the GO term must be between 5 and 50 (5 ≤ DE ≤ 50 genes). At each step, GO terms were replaced by “meta-terms” which were defined as the union of DE genes in those GO terms. Smaller meta-terms were then combined with larger meta-terms. The process continued iteratively until merges were no longer possible. The number of unique DE genes in each meta-term and its sub-terms were identified in an effort to understand changes in gene expression by major GO Biological Processes. Jaccard similarity was computed with describe the similarity in DE genes in meta-terms and sub-terms. Hub genes were identified using the meta-term assembly process. Within each sub-term, DE genes were ranked based on fold-change, and the DE genes exhibiting the greatest upregulation or downregulation were identified.

Gene-set enrichment analysis and enrichment mapping

DE genes with a log-fold change > 0.585 and a P value < 0.05 were carried forward for gene-set enrichment analysis (GSEA). Genes were ranked according to log-fold change and P value and GSEA was performed using 2,000 permutations and the March 2019 GO derived gene set from the Bader Lab (20). The enriched gene-sets were imported into the EnrichmentMap (EM) plugin in Cytoscape (version 3.7.1) with P value <0.001 and FDR-adjusted P <0.05 (21). Single nodes that did not map to a larger network were removed from the EM. Individual node labels were removed, and clusters were manually annotated for easier visualization.

Transcription factor analysis

Genomatix was employed to identify the relationship between DE transcription factors (TF) and the DE genes (Genomatix Software GmbH, Munich, Germany). For each gene-TF pair, Genomatix determined whether the TF has a validated or predicted binding site using interaction evidence from the literature. The algorithm computes four different strengths of interactions based on validated or predicted binding sites, evidence, and interactions with DE genes. We included TF with the strongest evidence of either a validated or predicted binding site and only report those TF that interacted with > 1% of the 800 DE genes. TFs were then cross-referenced in iPathway guide to identify their fold-change and FDR-adjusted P value.

Network analysis

DE genes and GO terms meeting the criteria described above were used to construct networks of key Hub Genes and Genes to GO terms in iPathway. The Hub Genes function was used to identify highly connected genes based on their degree of connectivity, influence, and gatekeeping. Genes with the greatest number of connections are placed in the center, and connections to other genes are coded based on the type of interaction. The genes to GO tool was employed to better understand how DE genes may affect specific GO biological processes.

RESULTS

Key differentially expressed genes

Eight hundred DE genes were identified out of a total of 9,118 genes with measured expression. The complete gene list with log fold change and P values is provided in table S1, https://links.lww.com/SHK/B90. Figure 1 shows the volcano plot, illustrating genes ranked based on the significance of their measured log fold change. The 10 greatest up- and down expressed genes based on log fold-change are listed in Table 1. Upregulated genes were involved in the promotion of cell division and proliferation (SYCP2, TC2N, PKP2, ITGA8, ENO1, HS3ST2), cell survival (TMEM74), membrane transport and GTPase activity (TBC1D8B), as well as dopamine modulation (OPRM1) and serotonin receptors (HTR1F). Downregulated genes were involved in the autophagy stress response (GALK2), cell motility (ACTC1), alkylglycerol cleavage (AGMO), neurodegenerative diseases (GSTO2, ASPA), tumor suppression (CDH19), cell senescence (CLCA2), cell cycle arrest (PPP1R1C), and cell adhesion (SGCD, GLB1).

F1
Fig. 1:
Volcano plot. Significantly differentially expressed (DE) genes (800) are represented in terms of their measured significance of the change (y-axis) and the expression change (x-axis). The significance is represented in terms of the negative log (base 10) of the P value. The dotted lines represent the thresholds used to select the DE genes: 0.58 for expression change and 0.05 for significance. The up-regulated genes (positive log fold change) are shown in red, while the down-regulated genes are blue. Not significant DE genes are shown in black.
Table 1 - Description of the 10 most upregulated and downregulated genes according to fold-change
Expression Symbol Gene name and ID Function Log fold-change P value
Up SYCP2 Synaptonemal complex protein 2 (10388) Increased expression is prognostic biomarker for shorter survival in breast cancer. Links homologous chromosomes during the prophase of meiosis. +2.36 9.219e-5
Up TC2N Tandem C2 domains, nuclear (123036) Oncogene that suppresses p53 signaling, and promotes cell division +2.09 0.006
Up PTCHD3 Patched domain containing 3 (374308) Unknown −2.015 0.014
Up OPRM1 Opioid receptor mu 1 (4988) Modulation of the dopamine system +1.98 0.004
Up TMEM74 Transmembrane protein 74 (157753) Promotes cell survival by inducing autophagy +1.89 0.005
Up PKP2 Plakophilin 2 Component of the desmosome and known for its role in cell-cell adhesion. Necessary to maintain transcription of genes that control intracellular calcium cycling; Promotes cell proliferation and migration via EGFR signaling +1.69 1.374e–4
Up ITGA8 Integrin subunit alpha 8 (8516) Facilitates phagocytosis of apoptotic cells, positive regulation of cell proliferation +1.58 0.011
Up TBC1D8B TBC1 domain family member 8B (54885) Functions as Rab-GAPs by binding to specific Rab proteins and affecting their GTPase activity. Rabs mediate different functions in membrane transport +1.53 1.834e–4
Up ENO1 Enolase 1 (2023) This gene encodes alpha-enolase, and functions plays a role in glycolysis and cell proliferation. +1.50 0.029
Up HS3ST2 Heparan sulfate-glucosamine 3-sulfotransferase 2 (9956) Member of the heparan sulfate biosynthetic enzyme family. Plays a role in cell proliferation +1.49 0.021
Up HTR1F 5-hydroxytryptamine receptor 1F (3355) Serotonin receptor 1F. Activation of serotonin receptors after TBI might decrease depression-like behaviors +1.47 0.014
Down GALK2 Galactokinase 2 (2585) Galactokinase activity at high galactose concentrations. Implicated in autophagy salvage pathway to produce glucose during stress −2.93 0.004
Down ACTC1 Actin, alpha, cardiac muscle 1 Plays a role in cell motility and cytoskeleton −2.50 0.011
Down AGMO Alkylglycerol monooxygenase (392636) Tetrahydrobiopterin- and iron-dependent enzyme that cleaves the ether bond of alkylglycerols. Implicated as cancer-promoting gene −2.35 0.017
Down GSTO2 Glutathione S-transferase omega 2 (119391) Enzyme involved in the metabolism of carcinogens and exnobiotics. Implicated in neurodegenerative diseases
Down CDH19 Cadherin 19 (28513) Calcium-dependent intercellular glycoprotein, might be a candidate tumor suppressor gene −2.06 0.013
Down PPP1R1C Protein phosphatase 1 regulatory inhibitor subunit 1C (151242) Major serine/threonine phosphatase that regulates various cell functions, including G2/M cell cycle arrest −1.94 0.022
Down SGCD Sarcoglycan delta (6444) Subcomplex of the dystrophin-glycoprotein complex (DGC), which forms a link between the F-actin cytoskeleton and the extracellular matrix −1.93 0.005
Down ASPA Aspartoacylase (443) Catalyzes the conversion of N-acetyl-L-aspartic acid (NAA) to aspartate and acetate. NAA helps maintain white matter. Low levels of NAA have been associated with in multiple sclerosis, epilepsy, and Alzheimer disease, and mutations associated with Canavan disease −1.93 0.041
Down CLCA2 Chloride channel accessory 2 (9635) Regulates chloride across the plasma membrane. Upregulated by p53 in response to DNA damage. Induces cellular senescence. Downregulated in breast cancer −1.85 0.008
Down GLB1 Gap junction protein beta 1 (2705) Encodes member of gap junction protein family. Mutations associated with X-linked Charcot-Marie-Tooth disease −1.80 0.030
Data obtained from the iPathway output.

Gene ontology analysis

Seven hundred ninety-one GO terms for Biological Processes were found to be significantly enriched based on uncorrected P values. The Elim pruning, threshold restrictions, and iterative merging procedures described above yielded 11 meta-terms (Table 2). The Signaling meta-term contains a high number of DE genes (150) mainly related to cell division, cell death, and the activation of transcription factors. For example, sub-terms involving the Pi3K, Ras, STAT, and TOR pathways were significantly modulated due to VPA treatment. Likewise, the Immune meta-term contained 160 DE genes due to VPA, such as those involved in the innate and myeloid immune pathways. Mitosis also emerged as a highly modulated meta-term, exhibiting 131 DE genes due to VPA treatment. The apoptosis meta-term revealed 81 DE genes due to VPA, with the intrinsic apoptosis pathway containing the greatest number of DE genes (53). Supplemental Figure S1, https://links.lww.com/SHK/B89 (to be electronically available on publisher's website) illustrates a denodgram based on the Jaccard similarity between GO meta-terms, sub-terms, and original GO-terms.

Table 2 - Meta-terms and sub-terms from the Gene Ontology Biological Processes iterative merging analysis by the University of Michigan Bioinformatics Core
Meta-term Sub-term Number of unique DE genes Upregulated genes Downregulated genes
Adhesion 33 PKP2, ITGA8, TEC, EPHA3, MPP7
Apoptosis 81
apoptosis.other 31 MSH2, CASP3, RFWD3, NR4A3, USP28 TRAF3IP2, NOX4, P2RX7, CTSZ, SIRT1
apoptosis.peptidase 23 ROBO1, CASP3, EFNA3, PIH1D1, CARD18 CAV1, ST18, PPARG, S100A8, F3
apoptosis.intrinsic 53 ENO1, MSH2, CASP3, ITPR1, CTH CAV1, MAL, ST18, PLA2R1, S100A8
Epigenetic 69
epigenetic.acetylation 49 CHAF1B, BCORL1, HMG20B, ATAD2, PIH1D1 SGF29, EPC1, SIRT1, PCGF1, BAZ2B
epigenetic.methylation 29 PIH1D1, TDRKH, SETD6, SMYD3, RTF1 APOBEC3H, ZFP57, SIRT1, THADA, SPI1
epigenetic.other 7 BRIP1, RPA2 TP53, INO80, MTERF1, MCM9, SMARCAL1
Immune 160
immune.humoral 17 IFNE REG1B, S100A8, TRAF3IP2, PSMB10, PTPN6
immune.Ifgamma 16 VPS26B, SLC26A6 HLA-DRA, RAB7B, KYNU, PPARG, CXCL16
immune.innate 59 MSH2, EREG, NR4A3, TEC, TICAM2 HLA-DRA, CAV1, RAB7B, IL1R1, PPARG
immune.lymphoid 27 MSH2, CASP3, IFNE, PTPRT, CYLD ST18, DOCK10, FLT3, PTPN2, BTK
immune.myeloid 68 EGR3, TESC, NR4A3, IL31RA, FUCA1 TMEM63A, RAB7B, PPARG, OLR1, S100AB
immune.other 15 PUS7, SIPA1L3, WDR7, SIN3A ANLN, FLT3, HERC6, SPI1, PTPN6
immune.pathogen 64 SKP2, AP1S3, DAG1, IFNE, SIN3A HLA-DRA, CAV1, REG1B, APOBEC3H, S100AB
Metabolic 101
metabolic.ATP 15 ENO1, DNAJC24, SLC25A12, PFKFB4, DNAJC1 P2RX7, THADA, TOR1AIP1, MYL4, DNAJB2
metabolic.carb 47 NEUROD1, PHKA2, NR4A3, EXT1, PIH1D1 GALK2, NOX4, P2RX7, MTMR2, SLC2A5
metabolic.lipid 14 NR4A3, PPARA, STARD4 PPARG, FLT3, P2RY12, TEK, TNFRSF1A
metabolic.other 45 PPARA, RAB27A, NME5, RNF144B, TBPL1 CAV1, APOBEC3H, NOX4, P2RX7, IL33
Mitosis 131
mitosis.chromatid_separation 61 SYCP2, USP44, RPRM, MSH2, EREG ANLN, GPNMB, EPC1, EVI2B, SLFN11
mitosis.error 46 USP44, CHAF1B, MSH2, SOX11, RFWD3 PROX1, SIRT1, TP53, INO80, EPC2
mitosis.DNA_replication 41 CHAF1B, MSH2, EREG, RFWD3, CDK2 PPARG, NOX4, SLFN11, SIRT1, TP53
mitosis.spindle 47 ULK4, GAS2L3, PSRC1, TRIM36, EPHA3 GAB1, INO80, GMFG, STAG1, TEK
ncRNA 42 MYBL1, PU57, PIH1D1, PPARA, TDRKH PPARG, NOP58, SIRT1, TP53, THADA
Organelle 64
organelle.other 33 ESYT3, CEP78, HAUS3, SFI1, ESYT1 ACTC1, MSR1, PPARG, PROX1, MTMR2
organelle.ER 38 TRAM2, TICAM2, RAB27A, RAB39A, BAIAP2L1 CAV1, RAB7B, P2RX7, CAV2, CTSZ
Signaling 150
signaling.growthFac 19 ROBO1, EGR3, NR4A3, MAPK14, SYCP2 APOD, GAB1, F3, FLT3, RAMP2
signaling.other 10 DAG1, BACE1 P2RX7, TNF, SPPL2A, HM13, SNX9
signaling.Pi3K 46 EREG, ROBO1, PROK2, PSRC1, SAMD5 NOX4, P2RX7, PROX1, FLT3, SIRT1
signaling.Ras 39 ROBO1, RND1, RRAGC, RACGAP1, RAB27A RAB7B, DOCK10, RTKN, RASGRP3, WASF2
signaling.STAT 62 MSH2, EREG, CASP3, SOX11, NEUROD1 CAV1, TNFSF18, NOX4, GPNMB, CLECL1
signaling.TOR 42 ENO1, NEUROD1, MINAR1, ITPR1, DEPDC5 CAV1, P2RX7, PTPN2, SIRT1, STAMBP
Transcription 18 HEY2, NR6A1, NR4A3, PIH1D1, PPARA PPARG, SGF29, TP53, ATF7IP, NR1H3
Transport 68
transport.lipid 31 ESYT3, PPARA, MAP2K6, ATP10D, ESYT1 CAV1, MSR1, PLA2R1, PPARG, SLCO1A2
transport.RNA 16 PIH1D1 GJB1, NOP58, SRSF6, P2RX7, EXOSC10
transport.other 22 USP44, SKP2, USP28, USP49, OTUB2 USP54, IL33, PSMB10, STAMBP, TP53
For each meta-term and sub-term, the number of DE genes and top five genes with the greatest upregulated or downregulated fold-change are listed. Threshold criteria included Elim-pruning correction, and ≥ 5 differentially expressed genes per GO term. GO terms were iteratively merged if they shared ≥ 95% of its constituent genes. Note that the sum of genes in a sub-term does not equal the number in its respective parent meta-term, as common genes may be present in > 1 sub-term.

Gene-set enrichment analysis and enrichment mapping

Of the 800 DE genes, GSEA produced 236 enriched gene-sets. Significantly up- and downregulated networks are shown in the enrichment map in Figure 2. Upregulated networks included neurotransmitter regulation and basal body cilium assembly, and major downregulated networks included those involved in the immune response, DNA transcription stress response, coagulation, cell motility, and ribosome and rRNA assembly.

F2
Fig. 2:
Enrichment map from gene set enrichment analysis (GSEA) illustrating networks of significantly enriched gene sets from the underlying analysis using Gene Ontology: Biological Processes. Node size represents the number of genes in the gene set; edge thickness is proportional to the overlap between gene sets; color intensity is proportional to the enrichment significance, with red representing upregulation and blue representing downregulation in FFP+VPA versus FFP groups, respectively. VPA indicates valproic acid; FFP, fresh frozen plasma.

Transcription factors

Genomatix analysis identified several TF that affected subsets of the 800 DE genes in the dataset. Table 3 lists TF with the number and percentage of DE genes they affect according to the literature (validated or predicted), log-FC, and FDR-adjusted P value. Major TFs that interacted with more than 10% of the total DE genes include tumor protein p53 (TP53), Myc proto-oncogene, peroxisome proliferator activated receptor gamma (PPARG), RELA proto-oncogene NF-kB subunit (RELA), and peroxisome proliferator activated receptor alpha. The TFs PPARG, hes related family basic helix-loop-helix (bHLH) transcription factor with YRPW motif 2 (HEY2), prospero homeobox 1, SRY-box 11 (SOX11), and neuronal differentiation 1 (NEUROD1), exhibited the greatest significance in terms of fold-change.

Table 3 - Transcription factors (TF) ranked by the percentage of DE genes that they affect
Transcription factor % of DE genes affected (total, validated, predicted) Log fold change FDR-adj P value
TP53 37.4%, (299 total, 26 validated, 273 predicted) −1.029 0.005
MYC 27.1% (217 total, 180 validated, 37 predicted) −0.891 0.013
PPARG 16.0% (128 total, 6 validated, 122 predicted) −1.419 0.004
RELA 10.4% (83 total, 55 validated, 28 predicted) −0.592 0.045
PPARA 10.0 (80 total, 0 validated, 80 predicted) +0.951 <0.001
AHR 8.3% (66 total, 0 validated, 66 predicted) +0.668 0.022
PPARD 8.0% (64 total, 0 validated, 64 predicted) +0.749 0.022
SNAI2 7.8% (62 total, 0 validated, 62 predicted) −0.830 0.043
SPI1 7.5% (60 total, 41 validated, 19 predicted) −1.004 0.004
SMAD1 6.6% (53 total, 0 validated, 53 predicted) −0.588 0.043
CTCF 5.9% (47 total, 41 validated, 6 predicted) −0.639 <0.001
CEBPD 5.3% (42 total, 17 validated, 25 predicted) −0.885 0.040
NEUROD1 4.5% (36 total, 0 validated, 36 predicted) +1.165 0.008
YBX1 4.4% (35 total, 0 validated, 35 predicted) −0.641 0.014
NR1H3 4.0% (32 total, 0 validated, 32 predicted) −0.702 0.024
PROX1 3.9% (31 total, 0 validated, 31 predicted) −1.183 0.044
PBX1 3.3% (26 total, 9 validated, 17 predicted) +0.660 0.031
EGR3 3.1% (25 total, 0 validated, 25 predicted) +1.113 0.044
IRF9 3.1% (25 total, 0 validated, 25 predicted) −0.778 0.030
HEY2 2.9% (23 total, 14 validated, 9 predicted) +1.416 0.004
SOX11 2.9% (23 total, 0 validated, 23 predicted) +1.173 0.003
Percentage, total DE genes affected, and the number of validated or predicted DE genes according to the literature are shown. Data were obtained from Genomatix analysis. Log fold-change and P values are for each TF.

Network analysis

Hub Gene analysis revealed a suite of seven genes that were highly interconnected and connected to additional DE genes (Fig. 3). TP53, C-X-C motif chemokine ligand 16 (CXCL16), and G protein subunit gamma 8 were downregulated with VPA treatment, while S-phase kinase-associated protein 2 (SKP2), HAUS augmin like complex subunit 3, adaptor-related protein complex 1 subunit sigma 3, and protein kinase cyclic AMP-activated catalytic subunit alpha were downregulated. Figure 4 shows key genes related to the innate immune response using the Genes to Go terms tool in iPathway, such as CXCL16, tumor necrosis factor (TNF), TNF receptor superfamily member 1A (TNFRSF1A), interleukin 33 (IL33), interleukin 1 receptor type 1 (IL1R1), and caveolin 1 (CAV1).

F3
Fig. 3:
Hub gene analysis showing clusters of highly connected subnetworks of interconnected genes using the Network Analysis tool in iPathway. The higher the number of connections, the closer to the center the gene will be drawn. Blue indicates a downregulated DE gene, red indicates an upregulated DE gene. A: activation, B: binding, C: catalysis, R: reaction. Data obtained from the iPathway output. DE indicates differentially expressed.
F4
Fig. 4:
Gene network illustrating key genes related to the innate immune response, showing how VPA affects genes related to the GO term “innate immune response” using the Genes to GO terms tool in iPathway. Blue: downregulated DE gene, Red: upregulated DE gene, Grey: measured, non-significant gene, White: not measured gene. DE indicates differentially expressed; GO, gene ontology.

DISCUSSION

This study aimed to understand the effects of combined administration of FFP and VPA on the brain transcriptome in a porcine model of TBI+HS. We focused on an area of the brain that was adjacent to the site of impact, as this penumbra is susceptible to secondary injury, but potentially salvageable with appropriate interventions. Our data revealed that FFP+VPA significantly modulated the transcription of hundreds of genes, biological processes, and networks involved in inflammation, neurogenesis, cell cycle regulation, and apoptosis. These findings build upon previous work demonstrating the ability of VPA to modulate gene expression (15, 16, 22, 23). Notably, those previous studies paired VPA with crystalloids (normal saline) or artificial colloids (Hextend), showing that the addition of VPA confers many cytoprotective benefits relative to fluid resuscitation alone. In this study, we sought to understand whether VPA could improve upon resuscitation with FFP alone, which is already widely considered to be one of the most beneficial resuscitation fluids for trauma patients. This study is the first to report that combined FFP+VPA resuscitation is associated with unique pro-survival changes in the brain transcriptome, beyond those invoked solely by FFP.

FFP, and to a lesser extent, VPA, have received considerable attention as next-generation treatment strategies for trauma. Plasma and its derivatives have been shown to improve survival in severely injured patients (24), and ongoing investigations are now elucidating the underlying mechanisms (8, 11, 25). Although the increased use of plasma-based resuscitation is an improvement over traditional crystalloids (24), we suggest that the addition of a pharmacologic agent, such as VPA, has the potential to even further reduce morbidity and mortality. In clinically realistic large animal models, VPA has been shown to improve survival from otherwise lethal injuries (26), reduce brain lesion size and swelling (9, 27), and improve neurologic recovery (10, 27). Recent evidence has shown that these clinical outcomes may be attributed to VPA's ability to alter gene expression and cell physiology (15, 16, 23, 28, 29), thereby creating a pro-survival phenotype (30–32). We have previously demonstrated that resuscitation with combined FFP+VPA resulted in a significant reduction of both brain lesion size and swelling compared with FFP alone (12). Follow-up investigations employed protein and cytological studies to demonstrate that FFP+VPA protects against cerebral mitochondrial dysfunction and excitotoxicity, as well as preserving platelet functions and protection of blood-brain-barrier integrity (30, 33). However, the present study is the first to reveal the impact of combined FFP and VPA treatments at the level of the brain transcriptome.

VPA attenuates the pro-inflammatory NF-kB pathway

TBI induces an acute inflammatory response that plays a key role in the pathogenesis of secondary brain injury (34). Local activation of glial cells, microglia, and astrocytes triggers the release of pro-inflammatory cytokines, ultimately resulting in swelling, edema, and weakening of blood-brain-barrier integrity. In this study, we found that VPA suppressed the inflammatory response following TBI+HS. Several cytokines and their receptors were downregulated, such as IL1R1 and IL33 (Fig. 4). We also found evidence suggesting that VPA attenuated inflammation via the NF-kB pathway. This potent inflammation pathway involves a multitude of signaling molecules and downstream effectors. TNF-alpha, in addition to initiating the extrinsic apoptotic pathway, also stimulates the production of CXCL16 (35), a pro-inflammatory cytokine and important hub gene in this analysis (Fig. 3). Downstream effectors include stimulation of tyrosine kinase binding protein (TYROBP) and RELA. The overall effect is potentiation of NF-kB signaling (36). In this study, each of these genes was downregulated in the VPA group, suggesting an overall attenuation of NF-kB signaling and inflammation. These findings are in line previous work by our group and others, who showed a downregulation of NF-kB signaling via the TYROBP-TREM2 complex (16) and attenuation of associated cytokine production (28, 37). Our lab has also shown that a single dose of VPA in addition to normal saline resuscitation can decrease neural apoptosis, inflammation, degenerative changes, and promote neural plasticity 30 days after TBI, by modulating the NF-kB/IkBα pathways (28). Importantly, however, our previous findings were in a model of TBI+HS that compared resuscitation using crystalloids or artificial colloid with and without VPA (20, 28). Taken together, the present findings and our previous work suggest that NF-kB pathway may be a common mechanism through which VPA reduces inflammation.

VPA potentiates neurogenesis

We found a suite of genes and TF involved in neurogenesis and neuronal differentiation that were upregulated in the VPA treated group (Table 3). NEUROD1 emerged as one of these key genes. This master TF regulates the transcription of a host of genes involved in neural differentiation during brain development, but is expressed at relatively low levels in adults (22). NEUROD1 is activated upon cellular stress, such as reactive oxygen species (ROS), in human embryonic stem cells (38). Additional genes and TF are known to play important roles in neurogenesis and neuronal differentiation, such as nuclear receptor subfamily 6 group A member 1 (NR6A1), fibroblast growth factor (FGF9), and SOX11. Each of the above molecules was upregulated in the VPA-treated group in the present study. Our findings support those of Higgins et al. (22) who showed that administration of VPA was associated with increased expression of these key genes, as well as NEUROD1 as a master TF. Nikolian et al. (23) likewise found an upregulation of both NEUROD1 and SOX11 in a similar model of TBI+HS using normal saline and a dose of VPA 50% lower than in the present study (i.e., 150 mg/kg). In addition, our lab has demonstrated that the effects of VPA on NEUROD1 and SOX2 can be measured in the RNA isolated from peripheral blood mononuclear cells, which creates the possibility of using them as biomarkers (29). Thus, a growing body of evidence suggests that VPA alters the brain transcriptome to promote remodeling.

VPA promotes cell cycle progression and limits cell senescence

In the context of trauma, enhancing cell cycle progression may beneficially promote cell division and tissue regeneration at the site of injury. Our data revealed that VPA was associated with the activation of pro-survival pathways while simultaneously inhibiting cell senescence and cell cycle arrest. Several key genes played central roles in a variety of signaling, cell cycle, and mitotic pathways. For example, we found an upregulation of genes involved in the G1-to-S phase checkpoint (cyclin-dependent kinase 2 and 8, and SKP2; Fig. 3). These genes were among the most overexpressed in our data, and played key roles in highly modulated GO terms involved in cell signaling and mitosis. At the same time, we found evidence of decreased cell cycle arrest and senescence. CAV1 and CAV2 are tumor suppressor gene candidates, eliciting a negative control on the Ras-p42/44 MAPK cascade (Table 2). Similarly, CLCA2 induces cell senescence via upregulation by p53, while PPP1R1C promotes G2/M cell cycle arrest (Table 1). The downregulation of these genes suggests attenuated cell senescence through multiple cellular pathways. Taken together, these findings indicate that VPA may enhance cell cycle progression while limiting senescence, with a net effect that may enhance neurogenesis in the injured brain.

TP53 inhibition may promote neuronal survival

TP53 repeatedly emerged as a key gene in our bioinformatic analyses. This important TF was not only significantly downregulated in the VPA group, but TF analysis indicated that it affected more than one-third of all differentially expressed genes (Table 3), making this TF the most interconnected gene in our study (Fig. 3). This tumor suppressor plays a central role in cancer, apoptosis, inflammation, autophagy, and metabolism. Cellular insults such as DNA damage or hypoxia activate TP53 to inhibit cell cycle progression and induce the intrinsic apoptotic pathway (39, 40). In this study, we observed decreased expression of TP53 in the VPA group, suggesting that VPA may suppress TP53-induced inflammation and apoptosis, and instead promote neuronal cell survival and proliferation. Emerging evidence suggests that TP53 activity may in part be regulated by its own acetylation status. Under homeostatic conditions in neurons, TP53 exhibits a steady state of acetylation (41). Neuronal stress disrupts this acetylation status to activate TP53 and initiate the intrinsic apoptotic pathway. However, administration of HDAC inhibitors restored acetylation status at specific residues on TP53 to inhibit its activation, attenuate apoptosis, and promote neuron survival. Thus, HDAC inhibitors, such as VPA, may regulate TP53 activity via two mechanisms: decreased transcription and restored acetylation.

VPA may promote the intrinsic apoptosis pathway while limiting necrosis

Apoptosis involves a coordinated, controlled mechanism of cell death. In this study, we found that VPA treatment may activate the intrinsic apoptotic pathway as evidenced by upregulated caspase 9 and caspase 3 (Table 2). The intrinsic pathway is induced via a variety of stressors (e.g., ischemia, ROS) that result in mitochondrial permeability, cytochrome c release into the cytoplasm, and activation of caspases (42). Interestingly, the present study suggests that TP53 did not play a major role in initiating this intrinsic pathway. Rather, endoplasmic reticulum (ER) stress, a potent activator of the intrinsic pathway, may play a role. Briefly, ER and mitochondria form mitochondria-associated membranes (MAM) for the direct exchange of calcium signaling and other materials. Mitochondrial calcium homeostasis is maintained by inositol 1,4,5-triphosphate receptor type 1 (ITPR1), a key receptor that controls ER calcium release across the MAM (Table 2) (43). As stressors such as ROS accumulate in the ER, ITPR1 initiates calcium influx into mitochondria, causing mitochondrial dysfunction resulting in disruption of ATP synthesis and membrane potential, and enhanced mitochondrial permeability (43). In this study, we found an upregulation of ITPR1 in the VPA group, suggesting increased mitochondrial calcium concentration and subsequent dysfunction. Thus, VPA may activate the intrinsic apoptotic pathway via this ROS-MAM axis.

In contrast, the extrinsic apoptotic pathway may be suppressed in the VPA group. A variety of death ligands are known to initiate the extrinsic pathway (42). Chief among these are tumor necrosis factor alpha (TNF-alpha) and TNFRSF1A, both of which were downregulated in the VPA group. The downregulation of this well-known cytokine–receptor complex suggests both an attenuation of inflammation and extrinsic apoptosis activity. Taken together, our results suggest that VPA may induce the intrinsic apoptotic pathway while attenuating the extrinsic pathway. While the promotion of apoptosis may initially seem like an unexpected outcome of VPA resuscitation, apoptosis may be far more beneficial than necrosis. Necrosis involves uncontrolled enzymatic degradation, massive intracellular leakage, and a disorganized and dramatic inflammatory response that exacerbates secondary brain injury. In contrast, apoptosis is an orderly, coordinated process where cellular components are disassembled, molecules are recycled, the cell membrane remains intact, and inflammation is minimized. Thus, induction of apoptosis due to VPA may be an important strategy to salvage precious neuronal tissue and limit inflammation in the penumbra following TBI. While we did not quantify the degree of necrosis in this study, ongoing histological and transcriptomic work in our lab is examining the relative roles of apoptosis versus necrosis in our large animal model of TBI+HS.

LIMITATIONS

This study has some limitations that need to be acknowledged. First, all animal models have limited direct translatability to human physiology and genetics. While imperfect, such models serve as an important platform to rigorously test emerging therapeutics. Second, our relatively small sample size increases the risk of a type 1 error. Increasing our sample size would decrease this risk, yet we feel that five animals per group strike the appropriate balance between scientific rigor and adhering to the “three R's” of animal research (replacement, reduction, and refinement). Additionally, the expense and labor of performing large animal experiments places practical limits on the sample size. However, this comes with the benefit or is more clinically realistic injury response in swine cortical impact models that is analogous to humans (44). Additionally, rodent models of TBI are limited in their translation to humans by dissimilar white to gray matter ratios and their lack of gyral complexity (45). Third, animals were euthanized 6 h after resuscitation (8 h after injury), yet brain lesion size and swelling peaks 48 to 72 hs after injury. We felt that transcriptomic changes at this early timepoint would be critical to capture while the penumbra is still salvageable.

CONCLUSIONS

This is the first study to show that VPA confers pro-survival epigenetic changes, even when compared to FFP. VPA-treated animals displayed significant increase in the expression of genes related to cell proliferation, neurogenesis, and the intrinsic apoptosis pathway, and decreased expression of pro-inflammatory genes in the injured brains. These findings add to the growing body of evidence supporting the unique neuroprotective properties of HDAC inhibitors.

Acknowledgments

The authors acknowledge generous support for translational research in traumatic brain injury by the Joyce and Don Massey TBI Foundation.

REFERENCES

1. Centers for Disease Control and Prevention. TBI data and statistics. Atlanta, Georgia, United States. 2014. Available at: http://www.cdc.gov/traumaticbraininjury/data/. Accessed April 18, 2019.
2. McMahon CG, Yates DW, Campbell FM, Hollis S, Woodford M. Unexpected contribution of moderate traumatic brain injury to death after major trauma. J Trauma 47:891–895, 1999.
3. Desai KH, Tan CS, Leek JT, Maier RV, Tompkins RG, Storey JD. Inflammation and the Host Response to Injury Large-Scale Collaborative Research Program. Dissecting inflammatory complications in critically injured patients by within-patient gene expression changes: a longitudinal clinical genomics study. PLoS Med 8 (9):e1001093, 2011.
4. Santry HP, Alam HB. Fluid resuscitation: past, present, and the future. Shock 33 (3):229–241, 2010.
5. Dekker SE, Sillesen M, Bambakidis T, Jin G, Liu B, Boer C, Johansson PI, Halaweish I, Maxwell J, Alam HB. Normal saline influences coagulation and endothelial function after traumatic brain injury and hemorrhagic shock in pigs. Surgery 156 (3):556–563, 2014.
6. Butler FK, Holcomb JB, Schreiber MA, Kotwal RS, Jenkins DA, Champion HR, Bowling F, Cap AP, Dubose JJ, Dorlac WC, et al. Fluid resuscitation for hemorrhagic shock in tactical combat casualty care: TCCC Guidelines Change 14-01—2 June 2014. J Spec Oper Med 14 (3):13–23, 2014.
7. Holcomb JB, Tilley BC, Baraniuk S, Fox EE, Wade CE, Podbielski JM, del Junco DJ, Brasel KJ, Bulger EM, Callcut RA, et al. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA 313 (5):471–482, 2015.
8. Halaweish I, Bambakidis T, He W, Linzel D, Chang Z, Srinivasan A, Dekker SE, Liu B, Li Y, Alam HB. Early resuscitation with fresh frozen plasma for traumatic brain injury combined with hemorrhagic shock improves neurologic recovery. J Am Coll Surg 220 (5):809–819, 2015.
9. Jin G, Duggan M, Imam A, Demoya MA, Sillesen M, Hwabejire J, Jepsen CH, Liu B, Mejaddam AY, Lu J, et al. Pharmacologic resuscitation for hemorrhagic shock combined with traumatic brain injury. J Trauma Acute Care Surg 73 (6):1461–1470, 2012.
10. Halaweish I, Bambakidis T, Chang Z, Wei H, Liu B, Li Y, Bonthrone T, Srinivasan A, Bonham T, Chtraklin K, et al. Addition of low-dose valproic acid to saline resuscitation provides neuroprotection and improves long-term outcomes in a large animal model of combined traumatic brain injury and hemorrhagic shock. J Trauma Acute Care Surg 79 (6):911–919, 2015.
11. Sillesen M, Bambakidis T, Dekker SE, Li Y, Alam HB. Fresh frozen plasma modulates brain gene expression in a swine model of traumatic brain injury and shock: a network analysis. J Am Coll Surg 224 (1):49–58, 2017.
12. Imam AM, Jin G, Duggan M, Sillesen M, Hwabejire JO, Jepsen CH, DePeralta D, Liu B, Lu J, deMoya MA, et al. Synergistic effects of fresh frozen plasma and valproic acid treatment in a combined model of traumatic brain injury and hemorrhagic shock. Surgery 154 (2):388–396, 2013.
13. Prevost TP, Jin G, de Moya MA, Alam HB, Suresh S, Socrate S. Dynamic mechanical response of brain tissue in indentation in vivo, in situ and in vitro. Acta Biomater 7 (12):4090–4101, 2011.
14. Jin G, DeMoya MA, Duggan M, Knightly T, Mejaddam AY, Hwabejire J, Lu J, Smith WM, Kasotakis G, Velmahos GC, et al. Traumatic brain injury and hemorrhagic shock: evaluation of different resuscitation strategies in a large animal model of combined insults. Shock 38 (1):49–56, 2012.
15. Dekker SE, Bambakidis T, Sillesen M, Liu B, Johnson CN, Jin G, Li Y, Alam HB. Effect of pharmacologic resuscitation on the brain gene expression profiles in a swine model of traumatic brain injury and hemorrhage. J Trauma Acute Care Surg 77 (6):906–912, 2014.
16. Bambakidis T, Dekker SE, Sillesen M, Liu B, Johnson CN, Jin G, de Vries HE, Li Y, Alam HB. Resuscitation with valproic acid alters inflammatory genes in a porcine model of combined traumatic brain injury and hemorrhagic shock. J Neurotrauma 33 (16):1514–1521, 2016.
17. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Statist Soc Ser B (Methodol) 57 (1):289–300, 1995.
18. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25 (1):25–29, 2000.
19. Alexa A, Rahnenführer J, Lengauer T. Improved scoring of functional groups from gene expression data by decorrelating GO graph structure. Bioinformatics 22 (13):1600–1607, 2006.
20. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102 (43):15545–15550, 2005.
21. Merico D, Isserlin R, Stueker O, Emili A, Bader GD. Enrichment map: a network-based method for gene-set enrichment visualization and interpretation. PLoS One 5 (11):e13984, 2010.
22. Higgins GA, Georgoff P, Nikolian V, Allyn-Feuer A, Pauls B, Higgins R, Athey BD, Alam HE. Network reconstruction reveals that valproic acid activates neurogenic transcriptional programs in adult brain following traumatic injury. Pharm Res 34 (8):1658–1672, 2017.
23. Nikolian VC, Dennahy IS, Higgins GA, Williams AM, Weykamp M, Georgoff PE, Eidy H, Ghandour MH, Chang P, Alam HB. Transcriptomic changes following valproic acid treatment promote neurogenesis and minimize secondary brain injury. J Trauma Acute Care Surg 84 (3):459–465, 2018.
24. Guyette FX, Sperry JL, Peitzman AB, Billiar TR, Daley BJ, Miller RS, Harbrecht BG, Claridge JA, Putnam T, Duane TM, et al. Prehospital blood product and crystalloid resuscitation in the severely injured patient: a secondary analysis of the prehospital air medical plasma trial. Ann Surg 2019; Online ahead of print.
25. Georgoff PE, Nikolian VC, Halaweish I, Chtraklin K, Bruhn PJ, Eidy H, Rasmussen M, Li Y, Srinivasan A, Alam HB. Resuscitation with lyophilized plasma is safe and improves neurological recovery in a long-term survival model of swine subjected to traumatic brain injury, hemorrhagic shock, and polytrauma. J Neurotrauma 34 (13):2167–2175, 2017.
26. Williams AM, Bhatti UF, Biesterveld BE, Graham NJ, Chtraklin K, Zhou J, Dennahy IS, Kathawate RG, Vercruysse CA, Russo RM, et al. Valproic acid improves survival and decreases resuscitation requirements in a swine model of prolonged damage control resuscitation. J Trauma Acute Care Surg 87 (2):393–401, 2019.
27. Nikolian VC, Georgoff PE, Pai MP, Dennahy IS, Chtraklin K, Eidy H, Ghandour MH, Han Y, Srinivasan A, Li Y, et al. Valproic acid decreases brain lesion size and improves neurologic recovery in swine subjected to traumatic brain injury, hemorrhagic shock, and polytrauma. J Trauma Acute Care Surg 83 (6):1066–1073, 2017.
28. Chang P, Williams AM, Bhatti UF, Biesterveld BE, Liu B, Nikolian VC, Dennahy IS, Lee J, Li Y, Alam HB. Valproic acid and neural apoptosis, inflammation, and degeneration 30 days after traumatic brain injury, hemorrhagic shock, and polytrauma in a swine model. J Am Coll Surg 228 (3):265–275, 2019.
29. Georgoff PE, Nikolian VC, Higgins G, Chtraklin K, Eidy H, Ghandour MH, Williams A, Athey B, Alam HB. Valproic acid induces prosurvival transcriptomic changes in swine subjected to traumatic injury and hemorrhagic shock. J Trauma Acute Care Surg 84 (4):642–649, 2018.
30. Nikolian VC, Dekker SE, Bambakidis T, Higgins GA, Dennahy IS, Georgoff PE, Williams AM, Andjelkovic AV, Alam HB. Improvement of blood-brain barrier integrity in traumatic brain injury and hemorrhagic shock following treatment with valproic acid and fresh frozen plasma. Crit Care Med 46 (1):e59–e66, 2018.
31. Williams AM, Dennahy IS, Bhatti UF, Biesterveld BE, Graham NJ, Li Y, Alam HB. Histone deacetylase inhibitors: a novel strategy in trauma and sepsis. Shock 52 (3):300–306, 2019.
32. Dekker SE, Nikolian VC, Sillesen M, Bambakidis T, Schober P, Alam HB. Different resuscitation strategies and novel pharmacologic treatment with valproic acid in traumatic brain injury. J Neurosci Res 96 (4):711–719, 2018.
33. Dekker SE, Sillesen M, Bambakidis T, Andjelkovic AV, Jin G, Liu B, Boer C, Johansson PI, Linzel D, Halaweish I, et al. Treatment with a histone deacetylase inhibitor, valproic acid, is associated with increased platelet activation in a large animal model of traumatic brain injury and hemorrhagic shock. J Surg Res 190 (1):312–318, 2014.
34. Hinson HE, Rowell S, Schreiber M. Clinical evidence of inflammation driving secondary brain injury: a systematic review. J Trauma Acute Care Surg 78 (1):184–191, 2015.
35. Abel S, Hundhausen C, Mentlein R, Schulte A, Berkhout TA, Broadway N, Hartmann D, Sedlacek R, Dietrich S, Muetze B, et al. The transmembrane CXC-chemokine ligand 16 is induced by IFN-gamma and TNF-alpha and shed by the activity of the disintegrin-like metalloproteinase ADAM10. J Immunol 172 (10):6362–6372, 2015.
36. Luo Q, Lin H, Ye X, Huang J, Lu S, Xu L. Trim44 facilitates the migration and invasion of human lung cancer cells via the NF-κB signaling pathway. Int J Clin Oncol 20 (3):508–517, 2015.
37. Li R, Aslan A, Yan R, Jongman RM, Moser J, Zwiers PJ, Moorlag HE, Zijlstra JG, Molema G, van Meurs M. Histone deacetylase inhibition and I(B kinase/nuclear factor-κB blockade ameliorate microvascular proinflammatory responses associated with hemorrhagic shock/resuscitation in mice. Crit Care Med 43 (12):e567–e580, 2015.
38. Hu Q, Khanna P, Ee Wong BS, Lin Heng ZS, Subhramanyam CS, Thanga LZ, Sing Tan SW, Baeg GH. Oxidative stress promotes exit from the stem cell state and spontaneous neuronal differentiation. Oncotarget 9 (3):4223–4238, 2017.
39. Aubrey BJ, Strasser A, Kelly GL. Tumor-suppressor functions of the TP53 pathway. Cold Spring Harb Perspect Med 6 (5):a026062, 2016.
40. Vousden KH, Prives C. Blinded by the light: the growing complexity of p53. Cell 137 (3):413–431, 2009.
41. Brochier C, Dennis G, Rivieccio MA, McLaughlin K, Coppola G, Ratan RR, Langley B. Specific acetylation of p53 by HDAC inhibition prevents DNA damage-induced apoptosis in neurons. J Neurosci 33 (20):8621–8632, 2013.
42. Muñoz-Pinedo C. Signaling pathways that regulate life and cell death: evolution of apoptosis in the context of self-defense. Adv Exp Med Biol 738:124–143, 2012.
43. Janikiewicz J, Szymański J, Malinska D, Patalas-Krawczyk P, Michalska B, Duszyński J, Giorgi C, Bonora M, Dobrzyn A, Wieckowski MR. Mitochondria-associated membranes in aging and senescence: structure, function, and dynamics. Cell Death Dis 9 (3):332, 2018.
44. Alessandri B, Heimann A, Filippi R, Kopacz L, Kempski O. Moderate controlled cortical contusion in pigs: effects on multi-parametric neuromonitoring and clinical relevance. J Neurotrauma 20 (12):1293–1305, 2003.
45. Xiong Y, Mahmood A, Chopp M. Animal models of traumatic brain injury. Nat Rev Neurosci 14 (2):128–214, 2013.
Keywords:

Epigenetic modulation; fresh frozen plasma; hemorrhage; histone acetylation; neuroprotection; neurotrauma; prosurvival strategies; valproic acid

Supplemental Digital Content

Copyright © 2020 by the Shock Society