The link between traumatic brain injury (TBI) and dementia may be unclear, but a new study suggests the possibility that the treatment for one may also suit the other. In a brief communication published online in yesterday's Nature Medicine, Mark Burns and colleagues at Georgetown University in Washington, DC, present evidence that shutting down the amyloid-producing enzymes β-secretase or γ-secretase leads to improvement in symptoms and reduction in neuronal death after head injury in mice. The results suggest that blocking the early activation of secretases may prevent later, secondary damage to brain tissue.

TBI is a risk factor for AD, and has been the focus of much press lately around the long-term consequences of sports-related concussions, especially among players in the NFL (see ARF live discussion on this topic). Aβ production ramps up rapidly after severe head injury, propelled by the induction of both secretases. Aβ plaques and signs of neuronal degeneration develop within days after TBI in one-third of patients (Ikonomovic et al., 2004). However, it is not known if or how those early events contribute to the long-term risk of AD, or even to the outcome of the initial injury.

To address the latter question, first author David Loane and colleagues looked at the effects of blocking secretase activation in mice that had suffered a direct impact to left parietal cortex. In their model, control mice showed an elevation of Aβ peptide within one day, which peaked at three days and returned to baseline by one week. APP, BACE1, and presenilin proteins also accumulated on the same schedule. The mice showed no dramatic problems with gross motor skills, but fine motor skills (walking on a beam) and hippocampal-based memory and learning (Morris water maze) were both affected.

When they created the same injury in β-secretase knockout mice, the accumulation of Aβ was non-existent. The mice showed better performance on the beam walking test within a week after injury compared to wild-type injured mice, and were comparable to uninjured mice in the water maze after 18 days. Likewise, giving wild-type mice a chronic administration of a γ-secretase inhibitor starting just after injury also caused a decrease in Aβ, which was down 25 percent compared to untreated mice after the last drug dose (at day 21). The mice improved in the walking test after three weeks’ treatment, and also showed a better performance in the water maze from one week on. MRI on control mice revealed extensive damage to the cortex, hippocampus, and lateral ventricles 21 days after injury. In the BACE1 knockouts, the extent of tissue damage was reduced by 30 percent, while the γ-secretase inhibitor DAPT reduced lesion volume by 70 percent. In both cases, tissue loss was decreased as assessed by histological staining, and staining of hippocampal neurons after injury revealed that the γ-secretase inhibitor partially blocked the loss of hippocampal neurons.

The outstanding question remains of whether the effects of secretase inhibition on injury outcome in the mice depend on blocking Aβ production, or whether other pathways might be involved. “At this time we are not 100 percent sure that it is going through Aβ,” Burns told ARF. “We can definitely say the secretases are playing a pivotal role.” Burns says he sees the secretases as stress proteins that increase levels and activity after trauma, and then have downstream effects, which may go through Aβ or not. “We know γ-secretase is involved in processing p75, Notch and Aβ, so you have potentially three different pathways that might account for some of the detrimental effects you see,” he said. The β-secretase is another story, Burns said, because we know little about other pathways downstream of the enzyme, leaving Aβ the prime candidate for its effects.

Arguing against an acute detrimental effect of Aβ itself is a recent paper from David Brody and colleagues at Washington University in St. Louis, Missouri (see ARF related news story), who showed that accumulation of Aβ in brain interstitial fluid in the first days after head injury paralleled neurological improvement. In that study, the increase in Aβ was interpreted to reflect increasing neuronal activity during recovery.

The scenario where the early activation of secretases after head injury acts act as a triggering event that leads to secondary tissue injury fits with the observation that the effects of secretase inhibition appear well after the peak of secretase activation: the researchers point out that secretase expression and Aβ production start soon after injury and peak after three days, but the effects of secretase inhibition on behavior only show up weeks after chronic treatment. Outcomes are immediately worse, then improve with time. “Once you start the progression to secondary injury, it’s like a switch,” Burns explained. “If you can stop that switch from activating, you have a chance to prevent downstream tissue loss and apoptosis.” The results suggest that early and acute treatment with γ- or β-secretase inhibitors in the hours or day after injury might help prevent the cascade of destruction, presenting a novel approach to the treatment for head trauma.—Pat McCaffrey


  1. I recommend the primary paper.

  2. The report from Loane et al. raises several intriguing questions:

    1. What are the mechanisms underlying the effects of DAPT treatment and BACE knockout in the setting of TBI?

    2. Can administration of APP secretase inhibitors starting at clinically realizable times after TBI be effective at improving cognitive and behavioral effects of injury?

    3. What pharmacodynamic assays for the most relevant effects of such secretase inhibitors should be used in preclinical and clinical studies?

    Hopefully, this report will stimulate further work on the important topic of the relationship between TBI and AD.

  3. This paper is on how blocking Aβ production may help treat traumatic brain injury. With respect to β-secretase, this was studied using BACE1 knockout mice, but with respect to γ- secretase, an inhibitor (DAPT) was used. The results look promising and suggest another important use for γ-secretase inhibitors. In this case, since chronic treatment may not be necessary, the Notch issue may not be so important (i.e., global inhibitors may be okay without the need for a Notch-selective modulator).

    View all comments by Michael Wolfe

Make a Comment

To make a comment you must login or register.


News Citations

  1. Soluble Aβ—Bane or Boon? Real-time Data in Humans Yield New Insight

Paper Citations

  1. . Alzheimer's pathology in human temporal cortex surgically excised after severe brain injury. Exp Neurol. 2004 Nov;190(1):192-203. PubMed.

Other Citations

  1. ARF live discussion

Further Reading

Primary Papers

  1. . Amyloid precursor protein secretases as therapeutic targets for traumatic brain injury. Nat Med. 2009 Apr;15(4):377-9. PubMed.