The aim of this study was to investigate the effects of the α2-adrenoreceptor agonist dexmedetomidine on cell survival in an established in vitro model for TBI. Using organotypic hippocampal slice cultures subjected to a focal mechanical trauma we found that dexmedetomidine had a dose-dependent protective effect on hippocampal cells. The dose-effect curve was U-shaped, with a concentration of 1 μM showing the strongest effect. Moreover, we observed neuroprotection by dexmedetomidine even when applied with a delay after the onset of the traumatic injury. Apparently, the neuroprotective effects of dexmedetomidine are mediated - at least in part - by ERK as they were abolished by the co-administration of PD98059. Interestingly, the co-application of hypothermia and dexmedetomidine did not exert synergistic effects.
We treated experimental groups with varying concentrations of dexmedetomidine after trauma. Results show that a concentration of 1 μM dexmedetomidine was most effective at reducing trauma intensity. This finding is consistent with data from a recent study using hippocampal slice culture subjected to oxygen and glucose deprivation [5
]. Further dilution as well as a higher concentration of dexmedetomidine provided lower protection until finally, at 100 μM and 0.01 μM, no difference compared to the positive control group was observed. Comparable observations have been made in other studies investigating dexmedetomidine's protective properties [5
Dexmedetomidine's neuroprotective properties have largely been attributed to its agonist actions at α2
]. Although recent investigations point out that its interaction with imidazoline I1 receptors and the activation of ERK might also play a role [5
] we did not explicitly analyze imidazoline I1
receptors. However, the results of our experiments indicate that activation of ERK is an important factor in the protection of traumatized nervous tissue by dexmedetomidine. ERK is a key enzyme in cell metabolism activated by many different types of tissue injury and has been attributed a "survival"- function. The inhibition of a survival signals itself can cause cells to undergo apoptosis; therefore application of PD98059, inhibitor of ERK's direct activator MEK1, might as well have had a negative effect on cellular survival that superposed with the positive effect of dexmedetomidine without specifically counteracting dexmedetomidine. To rule out this possibility western blots could have been performed. Comparison of levels of activated ERK for the positive control group, the 1 μM dexmedetomidine group and the 1 μM dexmedetomidine + PD98059 group could have backed up our hypothesis. This has recently been performed in a similarly designed study investigating dexmedetomidine's neuroprotective properties using organotypic hippocampal slice cultures subjected to oxygen and glucose deprivation. Results of this study showed that dexmedetomidine's and PD98059's effects do not simply superpose but that both act as two opponents on the same cascade [5
]. We therefore decided not to perform these tests and interpreted our finding as a sufficient hint towards an involvement of ERK in the mediation of the brain protective properties of dexmedetomidine in TBI.
Many experimental studies have revealed neuroprotective properties of hypothermia [13
] and the advantages of hypothermia in certain clinical circumstances have been demonstrated [24
]. To put our findings into context we subjected a group of slices to mild hypothermia of 32°C from trauma until final imaging. As expected, hypothermia alone did prove to be strongly protective, but 1 μM dexmedetomidine was even more effective at reducing trauma intensity. Interestingly, as described before in in vivo
models of incomplete cerebral ischemia, a combination of hypothermia and 1 μM dexmedetomidine did not result in synergistic effects [7
In a clinical setting, unpredictability is in the nature of TBI and application of any specific therapy will inevitably be delayed. Therefore we investigated the effect of dexmedetomidine when applied 2 or 3 h after trauma. In our model dexmedetomidine proved to be even more efficient when application was delayed by 2 h. However, by delaying application for 3 h no appreciable difference could be observed any more compared to immediate application.
We acknowledge that our results should be interpreted within the context of several limitations.
Propidium iodide labels the DNA of any disrupted cell and we were therefore not able to distinguish between necrosis and apoptosis. However, we aimed to quantify the total amount of cell death and it is agreed that propidium iodide uptake correlates well with the number of damaged cells [15
We opted for organotypic slice cultures because this model allows easy access to in vitro
manipulation of nervous tissue and yet mimics closely the in vivo
state of the tissue with respect to morphological and functional characteristics [26
]. Concerning the extrapolation to an in vivo
situation, results from organotypic slice cultures have been demonstrated to be a satisfying compromise between dissociated cell culture and in vivo
models using whole animals [28
]. We admit, however, that extrapolation of our results to the in vivo
situation is complex. In a living organism, the development and the outcome from TBI is significantly affected by a plethora of different variables that can only partly, if at all, be taken into consideration in an in vitro
model (e.g. cerebral perfusion pressure, edema of surrounding tissue, focal or global cerebral ischemia). Such conditions may be affected by dexmedetomidine independently from its direct effects on neuronal survival. Despite these disadvantages we believe that our model is appropriate as it focuses on the mechanical component of the injury and allows the analysis of the intrinsic neuroprotective properties of a drug independently from its possible interactions with confounding variables that can only poorly be controlled in an in vivo