Abstract
Bipolar disorder (BD) is a dysregulation of emotion and instability of mood, associated with abnormal brain network oscillations. BD is treated primarily with Li+, as well as with selected anti-epileptic drugs that can stabilize brain networks by acting on membrane ion channels. Despite the numerous years that have elapsed since Li+ was first discovered as a mood stabiliser, we still only have a vague idea of how Li+ exerts its mechanism of action. Most explanations favour intracellular secondary signalling mechanisms, which are likely to be involved either directly or indirectly. However, I hypothesized that additional meaningful information could be gleaned approaching Li+ treatment of BD as an electrophysiological and channelopathy issue. For example, Li+ may exert its effects on fundamental electrophysiological processes such as burst firing, which encodes, processes and propagates neuronal signals throughout the brain and is important in plasticity. Previous research focused on Li+ effects on action potential waveforms. Ion channel dynamics, however, also influence burst firing. For this project, I focused on how burst firing was affected by Li+ and how Li+ interacted with the extracellular milieu and the history of activity to alter burst firing.
I performed in vitro extracellular recordings on the major output neuron of the mouse Olfactory Bulb (OB) called the Mitral cells (MCs). I used the OB, as it is a well-characterized and highly conserved neuronal circuit, allowing greater translatability into neuronal processing in human sensory circuits. I used a specific stimulation paradigm of low frequency (0.1 Hz) single stimulation to the Olfactory nerve layer (ONL). The frequency of stimulation was low to allow me to observe both evoked and spontaneous activity, as well as to minimize the likelihood of inducing alterations in firing. I also used two brief epochs of delta-stimulation (DS) to the ONL (with and without Li+); Delta was used because it matched the breathing frequency of rodents to synchronise OB activity.
I measured the variance of the spikes (CV) and several burst activity metrics (duration, frequency, event count within burst) to see how Li+ influenced not only spike activity but burst firing.
In Chapter 1, I examined how altering the extracellular Ca2+ levels impacted neuronal firing in Li+. This was motivated by the hypothesis that individuals with BD may experience disrupted Ca2+ balance (discussed in Introduction below).
Interestingly, Li+ (after DS) significantly altered the burst frequency in an extracellular Ca2+-dependent fashion, and I also discovered a phenomenon where DS in Li+ (2mM) and low Ca2+ (0.7mM) abolished both evoked and spontaneous activity, inducing what looked like a depolarisation block.
The changes in MC burst activity would emerge after DS in Li+, suggesting an activity dependent effect of Li+. Thus, in Chapter 2 I removed the presence of the continuous 0.1Hz stimulation in Li+ (1 & 2mM, low Ca2+) to make sure the history of activity was not causing a priming effect. In Li+ (1mM) without continuous 0.1Hz stimulation, Li+ changed the qualitative properties of the MC firing, shifting it from phasic to tonic firing. This mode switching has previously not been reported as an effect of Li+ but has key implications for its effect on brain activity.
From Chapter 2 we knew that history of stimulation is important in the effect of Li+ on MC burst firing. In Chapter 3 to ensure that the delta-stimulation was not causing a priming effect (as each cell served as its own control). I conducted reverse order control experiments and repeated DS in Li+ (in low Ca2+) to see if it could replicate the abolishment of activity seen in Chapter 1, reversing the order of the experiment caused a decrease in the burst frequency which persisted even after washing Li+ out and the double DS in Li+ had no significant effect on MC activity.
In Chapter 4 I repeated the original experimental paradigm but exchanged Li+ for a low concentration of 4-AP (10 μM, a selective block for Kv1 channels like the delay current). I did this to determine if blocking of K+ channels by Li+ could be mediating the previously observed changes in burst firing. In physiological Ca2+ (1.3mM), DS in 4-AP increased the burst frequency (non-significantly), which is comparable to Li+ in 1.3mM Ca2+. However, in low Ca2+, 4-AP (unlike Li+) still (non-significantly) increased the burst frequency. Li+ may exert some effects by blocking Kv1 channels; however, this cannot account for its extracellular Ca2+-dependent effects.
Finally, in Chapter 5 I repeated the experimental paradigm in the tufted cell (TC) instead of the MC. This is because the TC plays a role in initiating and maintaining evoked and spontaneous burst firing in the MC, and I wanted to confirm whether the observed changes are due to alterations in the MC burst firing, not the TC. The TC burst firing was significantly different to the MC and did not change in response to DS or Li+, nor both in conjunction.
My research has shown that Li+ is exogenous stimulation-dependent and extracellular Ca2+ dependent, where it can change the directionality of effect dependent on the state of the history of activity and extracellular milieu. Additionally, Li+ can alter the firing mode in the bistable MCs (switching from bursting to tonic), which alters the signal processing and encoding. My electrophysiological research is unique as it looks at Li+ effect on MCs not over milliseconds but over an extended period (20min+). Moreover, I examined various extracellular Ca2+ concentration as would happen over time in different brain states, giving insight into how Li+ treats both mania and depression.