Much of the neuroscience research that is carried out in the School can be related to one or more diseases. Neurological conditions such as Alzheimer's disease, epilepsy and schizophrenia are the direct result of aberrant brain function while neuropathic pain results from problems in both the peripheral and central nervous systems. Changes in the central control of the sympathetic nervous system can lead to conditions such as diabetes, obesity and hypertension, which are closely linked to nuclei in the brainstem and thalamic structures while genetic conditions such as Down's syndrome have profound neurological effects. Research within the School impacts on all these conditions, either directly or indirectly.
There are many forms of dementia, but the best known and possibly most prevalent in Alzheimer's disease (AD). Over recent years, there has been a focus of work on the underlying changes in plasticity that is associated with this disease. Much of the work has focused on the role of the amyloidβ peptide (Aβ), which is produced by aberrant cleavage of the amyloid precursor protein (APP, reviewed in Randall et al 2010), using transgenic mouse lines carrying mutations that favour over-production of Aβ, namely TAS10 and Tg2576. Both of these mouse strains carry the swedish mutation of APP and both show accumulation of Aβ and the development of plaques. It has been shown that in aged animals, while there was reduced synaptic transmission and disrupted synchronous activity in the schaffer-collateral pathway in the hippocampus, there was no effect on LTP (Fitzjohn et al 2001, Brown at al 2005, Fitzjohn et al 2010). However, NMDA receptor independent LTP has recently been shown to be disrupted in the hippocampal mossy fibre pathway as well as short-term plasticity such as frequency facilitation (Witton et al 2010). Finding such deficits in plasticity in the hippocampus is important given the role played by the hippocampus in memory systems.
Other recent work with these mice has demonstrated the differential effect of amyloid plaque formation on the development of abnormal (dystrophic) neurites in different neuronal populations. Significantly more dystrophic neurites have been observed in cholinergic neurons than in galanergic neurons in the neocortex and hippocampus in aged Tg mice (Kelley et al 2011). This is compatible with the neuroprotective role of the neuropeptide galanin, which is known to be upregulated in Alzheimer's disease. In addition, we have also shown that mice that over-express this peptide (or a Gal2 receptor agonist) are less susceptible to Aβ1-42 toxicity than wild-type mice and that functional knockout of galanin or the Gal2 receptor increases cell death (Elliot-Hunt et al 2011). This raises the exciting possibility that Gal2 receptor agonists as a potential route to a treatment that could reduce the progression of symptoms in patients with AD.
Epilepsy is essentially an aberrant form of synaptic plasticity and the induction of epileptiform activity is amenable to study in brain slices in the same way as any other form of plasticity. Previous work has demonstrated the involvement of GluK1-containing kainate receptors (Smolders et al 2002), Group I mGlu receptors (Thuault et al 2003) and GABAB receptors (Brown et al 2003, Thuault et al 2005) in hippocampal epileptiform activity. More recently, collaborative work has shown that the A322D mutation in the α1 subunit of the GABAA receptor, which induces juvenile myoclonic epilepsy, results in increased endocytosis of these receptors (Bradley et al 2008). This would have the effect of reducing inhibitory capacity within the affected networks.
Further collaborative work has investigated the effects of chronic seizure activity on neuronal function in animal models of epilepsy. Repeated administration of 4-aminopyridine, generating daily seizures, results in a generalised increase in neuronal excitablility in the hippocampus but a decrease in the somatosensory cortex (Borbély et al 2009, Világi et al 2009). However, in both cases, an apparent rearrangement of AMPA receptor composition takes place, with a fall in the level of GluA2 subunits and a rise in GluA1 subunits, though there is an overall fall in AMPA receptor subunits. This may be expected to result in an increase in the number of Ca2+ permeable receptors. However, this is only apparent in dentate gyrus and stratum lacunosum-moleculare of the CA1 region of the hippocampus, as measured by Co2+ uptake (Borbély et al 2009). A fall in Co2+ uptake was seen in in somatosensory cortex (Világi et al 2009). Thus the increase an Ca2+ permeable receptors may underlie the increased excitability in the hippocampus and thus its susceptibility to recurrent seizures generated in the entorhinal cortex (Kopniczky et al 2005).
Schizophrenia, like all psychiatric conditions, is a complex disease that is thought to arise from the dysfuctional interaction of different brain regions. Processes such as learning, memory, attention and decision-making are emergent properties of co-ordinated, extended networks spanning cortical and subcortical structures and there is a growing consensus that GABAregic interneurons play an central role in the generation and maintenance of the network oscillations that are necessary to induce such co-ordinated activity (reviewed in Jones 2010).
Our work is aimed at understanding how neuronal firing in distant brain regions are related to such oscillations. For instance, not only is hippocampal place cell firing 'phase-locked' (i.e. firing is linked to a particular phase of the rhythmic pattern) to the 4-12Hz hippocampal θ rhythm but so is the firing of neurons in the pre-frontal cortex (PFC) during spatial learning tasks (Jones and Wilson 2005a). Indeed, the phase that the PFC neurons lock to changes with the behaviour of the animal (Jones and Wilson 2005b). Such phase-precession is co-ordinated across the PFC and occurs during decision-making processes. Disruption of the θ rhythm could lead to disorientation and confusion in navigating even a familiar landscape. Thus, it is easy to see how dysfunctional correlation rhythms can lead to serious cognitive deficits. We are now using a brain slice preparation that leaves the pathways between the hippocampus and PFC intact to further investigate the cellular and synaptic mechanisms that underpin such cross-correlation from one brain region to another.
In collaboration with colleagues at the Lilly Centre for Cognitive Neuroscience, we have recently used the well established E17 MAM model of schizophrenia to investigate the effects of NMDA receptor antagonists on EEG activity. NMDA open channel blockers, such as PCP and ketamine, produce psychotic symptoms and cognitive disturbances reminiscent of schizophrenia in healthy individuals and exacerbate symptoms in schizophrenic patients and we have now shown that they have regional-specific effects on γ rhythm and high-frequency oscillations (HFO) as well as inducing hyper-locomotion as well as increased wakefullness (Philips et al 2011). γ oscillations were increased across the brain, but the increase was much less marked in the visual cortex, whilst HFO was increased principally in the motor cortex. Further studies are required to further explore the EEG changes in this model and to look at the effects of both established and novel putative anti-psychotic agents. It is also hoped that the combination of behavioural, electrophysiological and anatomical methodologies together will help create better translation of drug action from rodent models to schizophrenic patients.
Pain generated by over-activation of peripheral sensory neurons either as a result of nerve injury or through disease (neuropathy) is a serious clinical problem. Research within the School is aimed at understanding the molecular basis of neuropathy. One of the major areas of research is the role of galanin. This neuropeptide is widely expressed in the brain and is upregulated following nerve injury, acting as a trophic factor in the central and peripheral nervous systems. (reviewed in Hobson et al 2008). Activation of the Gal2 receptor has been shown to have a neuroprotective effects in the hippocampus (Elliot-Hunt et al 2007) while galanin itself has recently been shown to play a role in inflammatory demyelination diseases such as multiple sclerosis (MS). Galanin was markedly upregulated in microglia in post mortem brain tissue from patients that had suffered from this disease, as well as in the experimental autoimmune encephalomyelitis (EAE) model of MS (Wraith et al 2009). Interestingly, galanin and Gal2 receptor knockout mice that had undergone EAE treatment developed disease symptoms quicker than wild-type mice, while transgenic animals in which galanin expression was rescued were completely resistant to the development of clinical disease. In addition, galanin acts as an antinociceptive, reducing allodynic behaviour even after it has become established (Pope et al 2010). These results, along with others, indicate that galanin and agonists for the Gal2 receptor could be promising avenues for the development of better therapeutic agents for the treatment of such neuropathic plain.
A further novel potential target for the alleviation of chronic neuropathic pain is PKMξ, an atypical variant of PKC. Collaborative work has recently shown that this kinase is upregulated in the anterior cingulate cortex following nerve injury and that a selective inhibitor, ZIP, blocked both synaptic potentiation and allodynia (Yao et al 2010).
A major role in the control of body weight homeostasis and obesity-induced hypertension has been established for the central melanocortin-4-receptor (Balthasar et al 2005). We are investigating key melanocortin-4-receptor-expressing neuronal subpopulations critical in the processes avoiding obesity-induced hypertension.
Another focus of research is the intracellular signaling cascades mediating and modulating neuronal gene transcription in the neuronal processes regulating energy homeostasis. Thus we have shown that the CREB-regulated transcription co-activator 2 (CRTC2) is expressed in several regions of the CNS, including the hippocampus as several nuclei in the hypothalamus (Lerner et al 2009). Furthermore, the sub-cellular distribution of CRTC2, as well as it's phosphorylation by AMPK, is regulated by glucose. A high glucose concentration was associated with high levels of nuclear expression rather than a cytoplasmic location; fasted mice showed increased phosphorylation over fed mice or fasted mice that had been given a glucose injection. Inhibition of AMPK led to translocation of nuclear CRTC2 from the cytoplasm to the nucleus, while activation showed the reverse re-distribution (Lerner et al 2009). Finally, CRTC2 was shown to be involved in the expression of the insulin receptor substrate 2 gene (Irs2), binding to the CRE domian of the promoter in a glucose-dependent manner, although the expression of another CRE gene, (Nuc2B) was unaffected. Thus CRTC2 is crucial in the regulation of a specific set of CRE genes that are involved in hypothalamic energy-sensing pathways.
The cerebellum is essential for the coordination of movements. Motor deficits of varying severity are common to individuals with Down's syndrome (trisomy 21), such as slower development of and difficulty in acquiring new motor skills, strabismus (squint), nystagmus (oscillating eye movements), impaired speech, and altered posture, gait and balance. In Down’s syndrome, there is loss of ~25% of granule neurons in the cerebellum. We are exploring how the intrinsic and synaptic properties of the surviving granule cells are altered in the Ts65Dn model.
One of the biggest challenges to understanding the causes of psychiatric disorders is the lack of good animal models to study the disease biology and potential new treatments. Researchers in Bristol are developing new methods to study the complex psychological processes involved in psychiatric disorders, such as emotion, perception and decision making, using animals. In order to enhance the translational validity of this work, and in collaboration with the School of Experimental Psychology, some of the methods developed for animals studies are also being tested in human volunteers and patients.