The nuclear factor erythroid 2-like 2 activator, tert-butylhydroquinone, improves cognitive performance in mice after mild traumatic brain injury

Abstract

Traumatic Brain injury affects at least 1.7 million people in the United States alone each year. The majority of injuries are categorized as mild but these still produce lasting symptoms that plague the patient and the medical field. Currently treatments are aimed at reducing a patient's symptoms, but there is no effective method to combat the source of the problem, neuronal loss. We tested a mild, closed head traumatic brain injury model for the effects of modulation of the antioxidant transcription factor Nrf2 by the chemical activator, tert-butylhydroquinone (tBHQ). We found that post-injury visual memory was improved by a 7 day course of treatment and that the level of activated caspase-3 in the hippocampus was reduced. The injury-induced memory loss was also reversed by a single injection at 30 min after injury. Since the protective stress response molecule, HSP70, can be upregulated by Nrf2, we examined protein levels in the hippocampus, and found that HSP70 was elevated by the injury and then further increased by the treatment. To test the possible role of HSP70, model neurons in culture exposed to a mild injury and treated with the Nrf2 activator displayed improved survival that was blocked by the HSP70 inhibitor, VER155008. Following mild traumatic brain injury, there may be a partial protective response and patients could benefit from directed enhancement of regulatory pathways such as Nrf2 for neuroprotection.

Fig. 1.

Experimental timeline. Thirty minutes following a mild closed head injury or sham injury, mice were treated with tBHQ by IP injection. Some mice were treated with multiple injections, which were administered daily for 6 days afterward for a maximum of 7 injections. Tissues were harvested at 1 and 3 days post injury for biochemical analysis. Behavioral analysis was performed at 7 and 30 days following injury.

Fig. 2.

Novel object recognition behavioral test. Injured and sham mice were treated with either vehicle or 33.4 mg/kg tBHQ for 6 days post injury and were analyzed for their ability to identify a “new” object (visual memory) at 7 days (n = 7–15) and (B) 30 days post injury (n = 6–9). The preference index equals (time near the “new” object − time near the “old” object)/(time near the “new” object + time near the “old” object). A significant decrease in preference index was observed in mTBI animals compared to sham animals at both 7 days (A) and 30 days (B) post injury. Treatment with tBHQ rescued this deficit 7 days post injury (A) and profound but no significant difference between the mTBI and mTBI + treatment groups could be observed at 30 days post injury. *P < 0.05.

Fig. 3.

Elevated plus maze behavioral test. Anxiety-like behavior was monitored for injured and sham mice, treated with either vehicle or 33.4 mg/kg tBHQ for 6 days post injury, at 7 days (n = 6–9) and 30 days post trauma (n = 9–10). Injured and sham mice treated with either vehicle or 33.4 mg/kg tBHQ for 6 days post injury were monitored for anxiety-like behavior at 7 (n = 6–10) post mTBI. No significant difference between groups was observed at 7 days post trauma.

Fig. 4.

Passive avoidance behavioral test. Injured and sham mice treated with either vehicle or 33.4 mg/kg tBHQ for 6 days post injury were monitored for their non-spatial learning ability. This test was performed 7 days (n = 6–8) following mTBI. No significant difference between groups was observed at days post trauma.

Fig. 5.

Activated caspase-3 expression following mTBI. Protein was harvested from the right hippocampus of injured and sham mice treated with either vehicle or 33.4 mg/kg tBHQ. Western blots were performed with antibodies specific to activated caspase-3. Expression was normalized to GAPDH. An increase in caspase-3 activation was observed in injured mice. Treatment with tBHQ reduced the activation of caspase-3 (n = 4–6, *P < 0.05).

Fig. 6.

HSP70 expression following mTBI. Protein was harvested from the right hippocampus and the right frontal cortex of injured and sham mice treated with either vehicle or 33.4 mg/kg tBHQ at 1 day or 3 days following injury. Western blots were performed with antibodies specific to HSP70 and expression was normalized to GAPDH. An increase in HSP70 protein was observed in the hippocampus of injured mice (n = 5–6) 3 days following injury and this was further elevated by the Nrf2 activator (*P < 0.01). No differences were observed at 1 day post injury (n = 4–5). No significant differences were observable in the cortex on 1 (n = 4–5) or 3 days (n = 5–6) post injury.

Fig. 7.

Viability of SH-SY5Y model neurons treated with tBHQ and HSP70 inhibitor. Undifferentiated human neuroblastoma SH-SY5Y cells were pretreated with vehicle, 10 μm VER155008, 10 μm tBHQ, or both VER155008 and tBHQ and subjected to multiple mild biaxial stretch injuries. Viability was measured 2 h following the injury by counting Annexin V-positive cells measured by confocal microscopy. A significant increase in Annexin staining was observed in injured cells. A decrease in Annexin staining was observed in injured cells treated with tBHQ compared to injured cells treated with vehicle (column 5 and 6). No significant difference in tBHQ treatment can be observed in the presence of HSP70 inhibitor (column 6 and 8). Data analysis was performed with ImageJ (sham n = 3, injured n = 6, *P < 0.01, **P < 0.001).

Fig. 8.

Novel object recognition test after single injection of treatment in vivo. Injured and sham mice treated with 1 injection of either vehicle or tBHQ (33.4 mg/kg) 30 min post injury and visual memory of a novel object was analyzed at 7 days (n = 9–18) and (B) 30 days post injury (n = 7–12). Preference index was calculated by (time near the “new” object-time near the “old” object)/time near the “new” object + time near the “old” object). A significant decrease in preference index was observed in mTBI animals compared to sham animals at both 7 days (A) and 30 days (B) post injury. Treatment of tBHQ rescued this deficit at 7 days (A) but not at 30 days (B) post injury. No difference in anxious-like behavior was observed between groups 7 days (C) or 30 days (D) post injury. No difference in non-spatial learning was observed between groups 7 days (E) or 30 days (F) post injury.