Altered adrenergic receptor signaling following traumatic brain injury contributes to working memory dysfunction

Abstract

The prefrontal cortex is highly vulnerable to traumatic brain injury (TBI) and its structural and/or functional alterations as a result of TBI can give rise to persistent working memory (WM) dysfunction. Using a rodent model of TBI, we have described profound WM deficits following TBI that are associated with increases in prefrontal catecholamine (both dopamine and norepinephrine) content. In this study, we examined if enhanced norepinephrine signaling contributes to TBI-associated WM dysfunction. We demonstrate that administration of α1 adrenoceptor antagonists, but not α2A agonist, at 14 days post-injury significantly improved WM performance. mRNA analysis revealed increased levels of α1A, but not α1B or α1D, adrenoceptor in the medial prefrontal cortex (mPFC) of brain-injured rats. As α1A and 1B adrenoceptor promoters contain putative cAMP response element (CRE) sequences, we therefore examined if CRE-binding protein (CREB) actively engages these sequences in order to increase receptor gene transcription following TBI. Our results show that the phosphorylation of CREB is enhanced in the mPFC at time points during which increased α1A mRNA expression was observed. Chromatin immunoprecipitation (ChIP) assays using mPFC tissue from injured animals indicated increased phospho-CREB binding to the CRE sites of α1A, but not α1B, promoter compared to that observed in uninjured controls. To address the translatability of our findings, we tested the efficacy of the FDA-approved α1 antagonist Prazosin and observed that this drug improves WM in injured animals. Taken together, these studies suggest that enhanced CREB-mediated expression of α1 adrenoceptor contributes to TBI-associated WM dysfunction, and therapies aimed at reducing α1 signaling may be useful in the treatment of TBI-associated WM deficits in humans.

Figure 1. Systemic administration of α₁ antagonists, but not α_2A agonists, improve WM performance in TBI animals

a) Rats were injured, then trained (on day 12 post-injury) in the WM version of the Morris water maze task. On day 14 post-injury, rats were injected with either the α₁ antagonist HEAT (0.1 mg/kg i.p.), the α_2A agonist guanficine (0.1 mg/kg i.p.) or vehicle, then tested for WM 30min later. b) Summary data showing the latency to locate the hidden platform in the location (Loc) and match trials (Match) in sham-operated control animals. c) Summary data showing the latency to locate the hidden platform in the location (Loc) and match trials (Match) during the last 3 trials of training for rats to be treated with HEAT (or vehicle). Rats treated with HEAT had improved WM performance when tested 30min post-injection. d) Summary data showing the latency to locate the hidden platform in the location (Loc) and match trials (Match) during the last 3 trials of training for animals to be treated with guanfacine (or vehicle). Rats treated with guanfacine remained impaired in the WM task when tested 30min post-injection. Significant interaction between trial number and treatment. *Significant difference in latency between location and match trial.

Figure 2. Brain injury increases the mRNA levels of α_1A adrenoceptor in the mPFC

Summary results of quantitative real-time PCR showing the fold-changes of mRNA levels of a) α_1A, b) α_1B and c) α_1D adrenoceptors in sham and brain-injured animals on day 14 after the surgery. *Significant difference between sham and brain injury. d) The levels of α_1A, but not α_1D, could be reduced by intra-mPFC infusions of Rp-cAMP. * Significant difference between vehicle and Rp-cAMP treatments.

Figure 3. CREB phosphorylation is increased in the mPFC on day 14 post- injury

a) Representative western blots for phospho-CREB and pan-specific CREB immunoreactivities in mPFC in sham and brain-injured animals on day 14 after injury. Summary results showing that b) the phosphorylation, but not c) the total levels, of CREB are increased in the mPFC following injury. *Significant difference between sham and brain injury.

Figure 4. Brain injury increases CREB phosphorylation in mPFC GABAergic neurons

Representative photomicrographs of double-immunostaings for a) phospho-CREB (brown) and NeuN (blue) and b) phospho-CREB (brown) and GAD67 (blue). Arrow, examples of double positive cells. Arrowheads, examples of mono-positive cells. Bar, 20μm. Summary results of the unbiased stereological cell counting for c) total phospho-CREB positive cells, d) phospho-CREB and NeuN double-positive cells, and e) phospho-CREB and GAD67 double positive cells. *Significant difference between sham and brain injury.

Figure 5. TBI increases the binding of phospho-CREB to the α_1A adrenoceptor promoter

a) Illustration of the promoter regions of α₁ adrenoceptors and the locations of CRE consensus sequences. b) Summary results of ChIP assay. Amplicons (CRE1, 2 or control) correspond to that shown in a). Results are shown as fold-changes in amplified target from each of the promoter regions relative to the amount detected in sham controls. *Significant difference between sham and brain injury.

Figure 6. Systemic administration of Prazosin on day 14 improves WM performance

a) Rats were injured, then trained (on day 12 post-injury) in the WM version of the Morris water maze task. On day 14 post-injury, rats were injected with the α₁ antagonist Prazosin or vehicle, then tested for WM at various time points. Summary data showing the latency to locate the hidden platform in the location (Loc) and match trials (Match) assessed b) 30min, c) 24hr, and d) 3 days after injection. Significant interaction between trial number and treatment. *Significant difference in latency between location and matching trial.

Figure 7. Hypothetical model for TBI-induced WM deficits

(1) TBI increases the activity of tyrosine hydroxylase (TH). This increased activity leads to (2) increased dopamine and norepinephrine content within the mPFC. (3) Gs-protein-coupled receptor stimulation leads to increased cAMP production and the activation of PKA. (4) Activated PKA translocates to the nucleus where it phosphorylates the transcription factor CREB. Increased transcription within GABAergic neurons leads to the (5) enhanced expression of α1 adrenoceptor, as well as GAD67 via a D1-dependent mechanism. (6) Enhanced α1 receptor expression and elevated GAD67 combine to depress mPFC activity and cause WM dysfunction.