Isoform 6-selective histone deacetylase inhibition reduces lesion size and brain swelling following traumatic brain injury and hemorrhagic shock.

BACKGROUND: Nonselective histone deacetylase (pan-HDAC) inhibitors, such as valproic acid (VPA), have demonstrated neuroprotective properties in trauma models. However, isoform-specific HDAC inhibitors may provide opportunity for more effective drug administration with fewer adverse effects. We investigated HDAC6 inhibition with ACY-1083 in an in vitro and an in vivo large animal model of injury. METHODS: Mouse hippocampal cells were subjected to oxygen-glucose deprivation (0% O2, glucose-free and serum-free medium, 18 hours) and reoxygenation (21% O2, normal culture media, 4 hours) with/without VPA (4 mmol/L) or ACY-1083 (30 nmol/L, 300 nmol/L). Cell viability was measured by methylthiazolyl tetrazolium assay. Expression of hypoxia-inducible factor-1α, heat shock protein 70, and effectors in the phosphoinositide-3 kinase/mammalian target of rapamycin pathway were measured by Western blot analysis. Additionally, swine were subjected to combined traumatic brain injury and hemorrhagic shock and randomized to three treatment groups (n = 5/group): (i) normal saline (NS; 3× hemorrhage volume); (ii) NS + VPA (NS; 3× hemorrhage volume, VPA; 150 mg/kg), and (iii) NS + ACY-1083 (NS; 3× hemorrhage volume, ACY-1083; 30 mg/kg). After 6 hours, brain tissue was harvested to assess lesion size and brain swelling. RESULTS: Significant improvement in cell viability was seen with both HDAC inhibitors in the in vitro study. ACY-1083 suppressed hypoxia-inducible factor-1α expression and up-regulated phosphorylated mammalian target of rapamycin and heat shock protein 70 in a dose-dependent manner. Lesion size and brain swelling in animals treated with pharmacologic agents (VPA and ACY-1083) were both smaller than in the NS group. No differences were observed between the VPA and ACY-1083 treatment groups. CONCLUSIONS: In conclusion, selective inhibition of HDAC6 is as neuroprotective as nonselective HDAC inhibition in large animal models of traumatic brain injury and hemorrhagic shock.

Rapid valproic acid-induced modulation of the traumatic proteome in a porcine model of traumatic brain injury and hemorrhagic shock.

BACKGROUND: Histone deacetylase inhibitors such as valproic acid (VPA) improve survival in lethal models of hemorrhagic shock and polytrauma. Although VPA is known to modulate transcription, its ability to reduce mortality within minutes of administration suggests involvement of a rapid, posttranslational mechanism. We hypothesized that VPA treatment would cause proteomic changes within minutes of treatment including quantitative and/or posttranslational differences in structural and/or effector proteins. MATERIALS AND METHODS: We used a porcine model of traumatic brain injury (computer-controlled cortical impact, 12 mm depth) and hemorrhagic shock (40% hemorrhage). Animals were kept in shock for 2 h and randomized to two groups (n = 3): normal saline (volume = 3:1 hemorrhage volume) or normal saline + VPA (150 mg/kg, single dose). Peripheral blood mononuclear cells were collected at baseline, postshock, and postresuscitation. Intracellular protein profiles were assessed using 1 dimensional gel electrophoresis, liquid chromatography, mass spectrometry, and analyzed with Ingenuity Pathway Analysis software. RESULTS: Animals treated with VPA demonstrated significant proteomic changes. Quantitative differences were found in over 200 proteins including effector, regulatory, and structural proteins in critical cell signaling pathways. Posttranslational modification analysis demonstrated differential VPA-induced acetylation of lysine residues in histone and nonhistone proteins. Pathway analysis correlated these changes with significant increases in numerous prosurvival and cytoskeletal intracellular pathways, including Rho GTPase signaling (P = 1.66E-11), integrin signaling (P = 4.19E-21), and a decrease in Rho guanosine nucleotide dissociation inhibitor signaling (P = 4.83E-12). CONCLUSIONS: In a porcine model of severe injuries, a single dose of VPA is associated with protective changes in the proteome that are measurable within minutes of treatment.

Valproic acid induces prosurvival transcriptomic changes in swine subjected to traumatic injury and hemorrhagic shock.

BACKGROUND: Valproic acid (VPA) is a histone deacetylase inhibitor that improves outcomes in large animal models of trauma. However, its protective mechanism of action is not completely understood. We sought to characterize the genetic changes induced by VPA treatment following traumatic injuries. METHODS: Six female Yorkshire swine were subjected to traumatic brain injury (controlled cortical impact), polytrauma (liver and splenic laceration, rib fracture, rectus crush), and hemorrhagic shock (HS, 40% total blood volume). Following 2 hours of HS, animals were randomized to resuscitation with normal saline (NS) or NS + 150 mg/kg of intravenous VPA (n = 3/cohort, 18 samples total). Blood samples were collected for isolation of peripheral blood mononuclear cells at three distinct time points: baseline, 6 hours following injuries, and on postinjury day 1. RNA was extracted from peripheral blood mononuclear cells and sequenced. Differential expression analysis (false discovery rate < 0.001 and p value <0.001) and gene set enrichment (Panther Gene Ontology and Ingenuity Pathway Analysis) was used to compare VPA to non-VPA-treated animals. RESULTS: A total of 628 differentially expressed RNA transcripts were identified, 412 of which were used for analysis. There was no difference between treatment groups at baseline. The VPA-induced genetic changes were similar at 6 hours and on postinjury day 1. Upregulated genes were associated with gene expression (p 2.13E-34), cellular development (1.19E-33), cellular growth and proliferation (1.25E-30), and glucocorticoid receptor signaling (8.6E-21). Downregulated genes were associated with cell cycle checkpoint regulation (3.64E-22), apoptosis signaling (6.54E-21), acute phase response signaling (5.84E-23), and the inflammasome pathway (1.7E-19). CONCLUSION: In injured swine, VPA increases the expression of genes associated with cell survival, proliferation, and differentiation and decreases those associated with cell death and inflammation. These genetic changes could explain the superior clinical outcomes in VPA-treated animals, including smaller brain lesion size and improved neurologic recovery.

Transcriptomic changes following valproic acid treatment promote neurogenesis and minimize secondary brain injury.

BACKGROUND: Early treatment with valproic acid (VPA) has demonstrated benefit in preclinical models of traumatic brain injury, including smaller brain lesion size, decreased edema, reduced neurologic disability, and faster recovery. Mechanisms underlying these favorable outcomes are not fully understood. We hypothesized that VPA treatment would upregulate genes involved in cell survival and proliferation and downregulate those associated with cell death and the inflammatory response. METHODS: Ten female swine were subjected to a protocol of traumatic brain injury and hemorrhagic shock. They were assigned to two groups (n = 5): normal saline (NS; 3× volume of shed blood), or NS + VPA (150 mg/kg). Following 6 hours of observation, brain tissue was harvested to evaluate lesion size and edema. Brain tissue was processed for RNA sequencing. Gene set enrichment and pathway analysis was performed to determine the differential gene expression patterns following injury. RESULTS: Animals treated with VPA were noted to have a 46% reduction in brain lesion size and a 57% reduction in ipsilateral brain edema. Valproic acid significantly upregulated genes involved in morphology of the nervous system, neuronal development and neuron quantity. The VPA treatment downregulated pathways related to apoptosis, glial cell proliferation, and neuroepithelial cell differentiation. Ingenuity Pathway Analysis identified VPA as the top upstream regulator of activated transcription, supporting it as a direct cause of these transcriptional changes. Master transcriptional regulator NEUROD1 was also significantly upregulated, suggesting that VPA may induce additional transcription factors. CONCLUSION: Administration of VPA attenuated brain lesion size, reduced brain edema, and induced significant changes in the transcriptome of injured brain within 6 hours. Patterns of differential expression were consistent with the proposed neurogenic and prosurvival effects of VPA treatment.

Educational content and the use of social media at US departments of surgery.

BACKGROUND: The growth of the social media platform Twitter has prompted many to consider its potential as an educational tool. Little is known about how surgery training programs are utilizing this resource and whether this platform can provide educational content effectively. We sought to determine national utilization of Twitter by departments of surgery in the United States and evaluate if educationally driven content heightened engagement with the Twitter followers. METHODS: We conducted a cross-sectional analysis of social media presence for all Accreditation Council for Graduation Medical Education accredited general surgery training programs between October 1, 2016 and December 31, 2016. Each tweet was characterized as either promotional or educational. Metrics related to account engagement, including impressions (number of times a tweet is seen) and retweets (number of times a tweet is shared), were compared. These results were compared against a single departmental account focused primarily on educational content. RESULTS: Thirty-two departmental Twitter accounts were identified from the 272 programs approached associated with accredited general surgery training programs. Training programs posted a median of 1.0 unique tweets (interquartile range: 0.6-2.3) per week. Tweets were primarily promotional (81% of posts) and generated marginal engagement with followers (3.4 likes/tweet; 1.5 retweets/tweet). In contrast, a single, resident-run departmental account at our institution (University of Michigan) focused on educational content generated consistent, educational content (19.6 unique tweets/week, 48% of which were educational), which resulted in increased engagement with followers (11.4 likes/tweet; 5.9 retweets/tweet) compared to other accounts. CONCLUSION: Though Twitter is being widely adopted widely by departments of surgery, it is primarily utilized for promotional content. Use of educational content may improve engagement from followers.

Valproic acid decreases brain lesion size and improves neurologic recovery in swine subjected to traumatic brain injury, hemorrhagic shock, and polytrauma.

BACKGROUND: We have previously shown that treatment with valproic acid (VPA) decreases brain lesion size in swine models of traumatic brain injury (TBI) and controlled hemorrhage. To translate this treatment into clinical practice, validation of drug efficacy and evaluation of pharmacologic properties in clinically realistic models of injury are necessary. In this study, we evaluate neurologic outcomes and perform pharmacokinetic analysis of a single dose of VPA in swine subjected to TBI, hemorrhagic shock, and visceral hemorrhage. METHODS: Yorkshire swine (n = 5/cohort) were subjected to TBI, hemorrhagic shock, and polytrauma (liver and spleen injury, rib fracture, and rectus abdominis crush). Animals remained in hypovolemic shock for 2 hours before resuscitation with isotonic sodium chloride solution (ISCS; volume = 3× hemorrhage) or ISCS + VPA (150 mg/kg). Neurologic severity scores were assessed daily for 30 days, and brain lesion size was measured via magnetic resonance imaging on postinjury days (PID) 3 and 10. Serum samples were collected for pharmacokinetic analysis. RESULTS: Shock severity and response to resuscitation were similar in both groups. Valproic acid-treated animals demonstrated significantly less neurologic impairment between PID 1 to 5 and smaller brain lesions on PID 3 (mean lesion size ± SEM, mm: ISCS = 4,956 ± 1,511 versus ISCS + VPA = 828 ± 279; p = 0.047). No significant difference in lesion size was identified between groups at PID 10 and all animals recovered to baseline neurologic function during the 30-day observation period. Animals treated with VPA had faster neurocognitive recovery (days to initiation of testing, mean ± SD: ISCS = 6.2 ± 1.6 vs ISCS + VPA = 3.6 ± 1.5; p = 0.002; days to task mastery: ISCS = 7.0 ± 1.0 vs ISCS + VPA = 4.8 ± 0.5; p = 0.03). The mean ± SD maximum VPA concentrations, area under the curve, and half-life were 145 ± 38.2 mg/L, 616 ± 150 hour·mg/L, and 1.70 ± 0.12 hours. CONCLUSIONS: In swine subjected to TBI, hemorrhagic shock, and polytrauma, VPA treatment is safe, decreases brain lesion size, and reduces neurologic injury compared to resuscitation with ISCS alone. These benefits are achieved at clinically translatable serum concentrations of VPA. LEVEL OF EVIDENCE: Therapeutic (preclinical study).