The central and peripheral nervous systems are complex constellations of nerve cells intricately connected and ultimately controlling motor, sensory and cognitive functions. Research in the Neuroscience Division is concerned with the cellular and molecular mechanisms by which these networks are formed during development and how they function in maturity to integrate multiple inputs and transmit information along neural pathways in the brain and in the organ systems which they control. An increased understanding of the processes underlying normal physiological function is an essential basis for determining how systems are perturbed during disease states.

In 2003, research into both the central and peripheral nervous systems was bolstered with the arrival of Professor Trevor Lamb to head the Visual Neuroscience Laboratory, Associate Professor Christian Stricker to head the Neuronal Network Laboratory and Professor David Hirst to head the Autonomic Neuroeffector Transmission Laboratory. These groups add new dimensions in the areas of visual processing in the retina, synaptic dynamics and gastrointestinal motility.

Calcium plays a crucial role in both pre and postsynaptic events and complex imaging techniques are being employed by many groups within the Division to investigate at high resolution the compartmentalization of intracellular signaling components. Success with major equipment grants in 2003 has facilitated the purchase of a digital imaging setup with deconvolution software and will further enable the purchase in 2004 of a confocal microscope for gastrointestinal and vascular studies.

Professor Caryl Hill, Head of Division

Blood Vessel Laboratory
Leader: Professor C Hill

Cellular coupling through gap junctions is an important feature of all arteries and veins. Gap junctions are intercellular channels, which are comprised of members of the connexin (Cx) family of proteins and which permit electrical and chemical coupling between adjacent cells. Thus gap junctions can coordinate cellular activity amongst vascular cells and facilitate vasoconstrictor and vasodilator responses to a variety of neural and non-neural stimuli.

Our research program is aimed at determining whether the expression and distribution of gap junctions and their component connexins varies between different blood vessels, whether such variation is reflected in differences in functional properties and whether connexins may represent a therapeutic target during vascular disease.

We have continued to amass strong evidence in favour of the hypothesis that gap junctional coupling of endothelial cells with smooth muscle cells through myoendothelial gap junctions (MEGJs) underlies the action of the important vasodilator factor, EDHF or endothelium derived hyperpolarizing factor. Using serial section electron microscopy and electrophysiology, we have shown a tight correlation between the incidence of MEGJs and the magnitude of EDHF-mediated vasodilation. Pharmacological studies have confirmed that in the caudal artery and the mesenteric artery of adult normotensive rats, there is no role for chemical factors and conversely that putative blockers of gap junctions abolish all vasodilator activity. We have found that the role of EDHF and the presence of MEGJs is greater in smaller vessels, suggesting that EDHF may play an important part in peripheral vascular resistance and hence blood pressure. We have also uncovered an additional role for EDHF during development, particularly in larger vessels, in which both EDHF and MEGJs are upregulated. A further degree of functional complexity has emerged from collaborative studies with colleagues at Yale University School of Medicine in that the conductance of gap junctions may be physiologically regulated.

We have also demonstrated that gap junctions are critical to rhythmical contractions which are observed in a number of arteries and arterioles both in vivo and in vitro. While these contractions rely on the release of calcium from intracellular stores in the vascular smooth muscle cells, it appears that the endothelium is essential for coordination of these calcium oscillations and that gap junctions are involved. Current experiments are focused on the role of particular connexins in these events.

Hypertension is a vascular disease characterized by endothelial dysfunction and an imbalance between vasoconstriction and vasodilation. We have found significant alterations in the incidence and distribution of connexins in both the smooth muscle and endothelial cell layers of several arteries and in the morphology of these vessels. These anatomical changes are correlated physiologically with a reduced coupling of smooth muscle cells in the media, although little change has been found in the activity of EDHF. Future experiments will determine whether the alterations in connexins in the endothelium may affect other vascular functions such as conduction of responses along the vessels.

The Synaptic Dynamics lab has seen a year of wide-ranging collaboration and undergraduate education. Four research projects have been completed in collaboration with groups based in Sydney and Germany, resulting in the submission of three manuscripts, and a fourth that is very close to submission. Another collaboration with colleagues in Melbourne and produced a review article in the prestigious journal 'Trends in Neurosciences'. Three honours students have successfully completed research projects in the lab, and graduated with excellent marks. All three have either begun a PhD or are planning to start one early next year. Their research results were of high quality and will form the basis of two manuscripts to be submitted next year.

Two projects focused on surprising and unexpected properties of the NMDA receptor – an intensively studied protein that is thought to underlie learning and memory. In collaborations with Professor David Adams and Dr Ania Kloda at the University of Queensland, we have shown that ATP has complex actions on NMDA receptors. At low concentrations of glutamate, ATP tends to suppress the activity of the NMDA receptors by competing for the glutamate binding site, but at higher concentrations of glutamate, ATP enhances the activity of the receptor via an allosteric mechanism. During normal synaptic transmission, the newly discovered allosteric enhancement is likely to predominate. In a separate collaboration with Dr Greg Stuart and Bjöern Kampa at the University of Freiburg, Germany, we have studied the block of NMDA receptors by magnesium.

Another collaboration focused on the glycine transporter. This protein may also play a role in regulating NMDA receptor activity, as glycine is a co-agonist at the NMDA receptor. Together with Karin Aubrey and Dr Rob Vandenberg at the University of Sydney, we have analysed the biophysical and kinetic properties of the GlyT1b transporter, which is found predominantly in glial cells surrounding neurons and synapses in the brain. We have defined the conditions under which GlyT1b may release glycine into the synaptic cleft via reverse transport. Such release may enhance NMDA receptor activity during learning, or under pathological conditions.

Glycine transporters also clear glycine from inhibitory synapses in the brain stem and spinal cord where glycine receptors predominate. The relationship between the structure and function of glycine receptors was probed using substituted cysteine accessibility analysis, in collaboration with Dr Joe Lynch and Reena Han at the University of Queensland.

The glycine receptor was also the focus of two research projects carried out here in the Synaptic Dynamics lab. A mutant form of the glycine receptor is known to underlie the rare inherited neurological disease, hyperekplexia (also known as 'startle syndrome'). An unexpected noise or physical contact can produce seizures in people who suffer from this disease, and it blights their lives. Gemma Derrick and Alice Tindall tested a wide range of drugs on a model system expressing glycine receptors with a hyperekplexia mutation. Unfortunately, the drugs we found which boosted the activity of the mutant receptor only did so at toxic concentrations. Nonetheless, the study may provide some guidance to future efforts to find a cure for this disease. Another study with Sung Eun Kwon focussed on glycine receptor structure and function. We discovered that the glycine receptor can be activated while the competitive antagonist, strychnine, is still bound to it. In fact the antagonist makes the receptor hyperactive under some conditions. This completely unexpected result changes the way we think about how agonists and antagonists interact at a receptor.

Cerebral Cortex Laboratory
Leader: Dr J Bekkers

Research in the Cerebral Cortex Laboratory focuses on the electrical properties of neurons in the higher brain regions of rodents, with the aim of answering basic questions about synaptic transmission and integration. We use patch clamp and imaging techniques in dissociated cell cultures, in brain slices, and 'in vivo' in intact brains. By studying the brain at different levels of reductionism, we hope to contribute to an understanding of how the healthy nervous system carries out its computations, and what goes wrong in disease states like epilepsy.

During 2003 we have worked on four main projects.

1. The effect of the dendritic tree on firing properties of neurons. The dendritic tree is generally regarded as being beneficial to a neuron, enriching its computational repertoire. However, our results show that dendrites may come at a cost: they impose an electrical load on the axon, making it more difficult for the neuron to generate an output of action potentials. This project was begun during my sabbatical last year in the laboratory of Professor Michael Häusser, University College London. At that time we established a new 'dendrotomy' technique for occluding or removing the dendritic tree while recording from the soma of a cerebellar Purkinje cell. This year we have extended the dendrotomy method to other types of neurons (notably, cortical pyramidal cells), and have further characterized the excitability changes. For example, we have used a dynamic clamp to examine the effect of dendrotomy on synaptic integration, and we have also studied changes in burst firing. In addition, we have been using the neural simulation program, Neuron, to shed light on our experimental findings. Some of this modelling was done during another brief visit to Häusser's laboratory in London in July/August, and suggested further experiments that I am currently doing in Canberra.

2. The role of syntaxin in neurotransmitter release. Neurotransmitter release is known to involve an array of specialized presynaptic proteins, but many questions remain about the role of each component. We are addressing this issue by overexpressing a non-native form of one of these proteins, syntaxin, in hippocampal neurons in culture, and then studying the electrophysiological consequences of this substitution. As a first step, we are setting up to record calcium currents in these neurons, in order to test the hypothesis that the different isoforms of syntaxin interact differently with calcium channels. This project is being done by a new Graduate Diploma student in the lab, Chris Cassella, and also involves a collaboration with Dr Catherine Morgans (Neurological Sciences Institute, USA).

3. Characterization of 'presynaptically-silent' GABA synapses in hippocampal cultures. In 2001 I first observed a new kind of 'presynaptically-silent' GABA synapse at autaptic connections on isolated hippocampal pyramidal neurons grown in culture. These synapses express postsynaptic GABA_A receptors but lack presynaptic GABA. This year I have completed an extensive series of experiments designed to characterize the properties of this synapse. The experiments confirm that a brief pulse of externally-applied GABA can be endocytosed into presynaptic vesicles. These vesicles also contain a normal concentration of the native neurotransmitter, glutamate. The co-packaged glutamate and GABA can subsequently be released by electrical stimulation, giving rise to a mixed postsynaptic EPSC/IPSC. Because pyramidal neurons do not normally synthesize GABA, the loaded GABA is steadily depleted with continued stimulation, causing a steady reduction in amplitude of the IPSC. The rate of decline of the IPSC reflects in part the kinetics of movement of vesicles through the synaptic vesicle cycle, providing a novel assay of this process. Future experiments will take advantage of this system to study the kinetics of vesicle cycling under different stimulation conditions.

4. Neurophysiology of a GalR1 knockout mouse with an epileptic phenotype. Galanin is a neuropeptide that is widespread in the central nervous system, but its exact function in the brain remains unclear. Drs Arie Jacoby and Tiina Iismaa at the Garvan Institute, Sydney, recently generated a knockout mouse that lacks the Type 1 Galanin receptor  (GalR1), and made the surprising discovery that these animals are spontaneously epileptic. Because rodent models of spontaneous epilepsy are rare, we would like to characterize the epileptic phenotype of these GalR1 knockout animals and establish the physiological basis for their seizures. This year I have continued a collaboration with Dr Craig McColl, a neurologist and PhD student with Professor Stephen Redman, JCSMR, who is an expert at making EEG recordings from freely-moving rodents. So far we  have found that virtually all of the mutant animals exhibit a much higher frequency of abnormal 'spike and slow wave' EEG patterns than control animals. We are currently setting up to make depth electrode recordings in order to better localize the site(s) of seizure onset. Meanwhile, I have made further patch clamp recordings in slices from mutant and control animals, focusing on brain regions that normally express high levels of GalR1. I have observed changes in inhibition which may underlie the seizure phenotype. We are currently following up this finding.

Developmental Neurobiology Laboratory
Leader: Professor I Hendry

Nerve cells are dependent on information they receive from the postsynaptic cell in order to survive and mature. One of the major problems for the survival of nerve cells is their extreme length, sometimes over a meter from the cell body along the axon to the terminal. Survival of neurons during development of the nervous system depends upon trophic factors signalling from the nerve terminals to the cell body. These factors act on receptors at the nerve terminal and are then faced with the problem of transducing their signal along the axon to the cell body. Understanding the processes that control the flow of information from the nerve terminal to the cell body is essential if we are to understand neuronal development and apply this understanding to enhance nerve regeneration. The results of our studies will suggest alternate ways to enhance neuronal regeneration by perturbing the second messenger cascades promoting axonal transport. The main players in this signaling process are a family of neurotrophic factors that include nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin 3 (NT-3) and neurotrophin 4 (NT-4). These factors promote the survival of sympathetic and sensory neuronal populations by binding to receptors present both in the cell body and on the nerve terminal. There are two classes of receptors, one is a low affinity neurotrophin receptor named p75^NTR, which binds all the neurotrophins. The other class consists of the high affinity tyrosine kinase receptors: TrkA that binds NGF, TrkB that binds NT-4 and BDNF and TrkC that bind NT-3. At the nerve terminal the neurotrophins bind to their receptors, and the neurotrophin-receptor complex is then internalised into a vesicle and transported to the cell body. While vesicular retrograde transport of neurotrophins in vivo is well established, relatively little is know about the mechanisms that underlie vesicle endocytosis and formation prior to transport from the varicosities. There are three different sites in the neuron where 2nd messenger proteins can interact with the signalling complex and be activated. Signaling cascades are initiated both at the nerve terminal and at the cell body when 2nd messengers are recruited to the plasma membrane by activated receptors. After receptor-mediated endocytosis, 2nd messenger molecules continue to be recruited to the internalised vesicle; however, the mix of proteins differs in the nerve terminal and in the cell body. At the nerve terminal this internalised NGF may be recycled back to the plasma membrane, or targeted for retrograde axonal transport within a signalling endosome, which includes other molecules to be retrogradely transported to the cell body. The mechanisms by which these processes are regulated are not fully understood. Potentially, the Rab family of proteins, known to characterise the steps of the endocytic pathway, may be involved in the internalisation, targeting and transport of NGF. The associations between Rab 4, 5a, 5b, 7 and 11, and NGF, were studied using antibodies against the Rab proteins. Fluorescently labelled biotinylated NGF was used to positively identify the retrogradely transported, NGF containing organelle in the sciatic nerve and SCG. It was found that Rab 4, 5a, 5b and 11 were retrogradely transported in the sciatic nerve and associated with NGF in the superior cervical ganglion sciatic nerve. Rab 7 was not retrogradely transported but associated with NGF in the SCG. When the retrograde neurotrophin signalling endosome reaches the cell body, it can recruit additional 2nd messenger molecules to finally generate the unique signal derived from the nerve terminal. The multivesicular body observed in vivo functions as an endosome carrier vehicle or retrosome. This retrosome enables the mix of signalling molecules recruited at the terminal to be transported intact to the cell body. This will allow the cell body to receive a snapshot of the events occurring at the nerve terminal at the time the retrosome is formed.

The GTP-binding protein, G_z can couple to a number of receptors including the μ opioid, serotonin 1A and dopamine receptors. We have generated a mouse, which is deficient in its α subunit gene, Gα_z,. The deletion of this gene results in a complex set of changes in the behavioural response of the mouse to a number of drugs. Testing for analgesia in response to heat or cold showed that Gα_z knockout mice develop tolerance to morphine more rapidly and to a greater extent than control mice. In these mice the increase in tolerance is not due to receptor down regulation, as the maximum binding and receptor affinity do not differ between control and knockout. Tolerant Gα_z knockout mice survived higher doses of morphine than wild type mice. Gα_z knockout animals showed significantly enhanced morphine induced locomotor activation. Tolerant Gα_z knockout animals showed a reduced naloxone precipitated jumping behaviour, suggesting dissociation between tolerance and dependence in these mice. The alteration in tolerance observed in these experiments is likely to be due to a perturbation of G-protein coupled second messenger cascades and G_z may normally play a role in the prevention of the development of opioid tolerance. These results may help to explain some of the variability in the response of patients to opiate pain therapy.

The plasma corticosterone level in response to morphine was elevated in Gα_z knockout animals. Cocaine and amphetamine-induced locomotor activity is significantly elevated in Gα_z knock-out mice. These motor stimulating effects are mediated by activation of central dopamine receptors coupled to G proteins. The dopamine D2-like receptor agonist quinpirole suppressed locomotor activity in both groups of mice, but this was smaller in Gα_z deficient mice. Quinpirole inhibition of dopamine release in the forebrain nucleus accumbens evoked by electrical stimulation of dopamine axons was significantly attenuated in mice lacking Gα_z. Hypothermia and adrenocorticotropic hormone (ACTH) release resulting from activation of dopamine D2-like receptors were also significantly reduced in Gα_z deficient mice. Whole-cell recordings of CA1 pyramidal cells in hippocampal slices prepared from pentobarbitone-anaesthetised mice showed that 5HT produced an outward current that was 80% greater in Gα_z knock-out mice than in control mice, and was abolished by the 5HT1A receptor antagonist, WAY-100635. Overall, the data provides the first evidence that G_z proteins are functionally coupled to dopamine D2-like receptors and 5HT1A receptors in vivo.

Movement and Memory Laboratory
Leader: Professor S Redman

The Neuronal Signalling Laboratory conducts basic research into how individual nerve cells in the brain process information. This work involves recording activity from single neurons using both electrical and optical techniques. During the course of the last year our research has focused on the following issues:

The distribution and properties of voltage-gated sodium channels

Voltage-gated sodium channels play an essential role in the generation and propagation of nerve impulses. We have conducted experiments aimed at determining the distribution and properties of these channels in neurons. In contrast to previous observations, we have found evidence based on electrophysiological, immuno-cytochemical, and fluorescent imaging experiments that indicates that the density of voltage-gated sodium channels is high in the axon initial segment of pyramidal neurons in the cortex. Further, we show that a high density of sodium channels in the axon initial segment is require for faithful propagation of nerve impulses from their axonal site of generation back to the soma and dendritic tree. This so-called “backpropagation” is thought to be essential for many forms of neuronal computation including learning and memory.

Cholinergic modulation of dendritic excitability

Cholinergic inputs to the cortex are thought to have an important role in modulating cortical processing, such as the facilitation of oscillatory activity, increasing neuronal excitability, and enhancing synaptic plasticity. We have investigated the effect of cholinergic activation on the processing of information in the dendrites of cortical pyramidal neurons. We find that cholinergic activation depolarises both the soma and dendrites of neurons, and influences the ability of nerve impulses to propagate back into the dendritic tree. These findings may account for some of effects of cholinergic activation on learning and memory.

Dendritic mechanism underlying synaptic plasticity

Most of the input to neurons is made onto the dendritic tree, which is thought to be the site of changes in synaptic strength underlying learning and memory. We have investigated the cellular mechanism underlying long-term changes in synaptic strength during pairing of pre- and postsynaptic activity. We find that activation of NMDA receptors by backpropagating nerve impulses, thought to be critical for some forms of learning and memory, has a slow component due to “trapping” of magnesium ions within the pore of the NMDA receptor channel. This finding has important implications for setting the “time window” for induction of changes in synaptic strength during pairing of pre- and postsynaptic activity.

Together, this research increases our understanding of how our brains work, and in particular the cellular mechanisms that neurons use to make memories. In the long run this should help in the development of therapies to treat conditions associated with memory loss, such as Alzheimer's disease.

Visual Neuroscience Laboratory
Leader: Professor Trevor Lamb

The Visual Neuroscience Laboratory was set up in 2003, upon the arrival of Professor TD Lamb in the Division. The overall aim of the group's research is to provide a detailed understanding of the molecular steps involved in the first stage of vision; that is, the conversion of light into a neural signal in the rod and cone photoreceptors of the retina. This is being tackled on the following fronts: by recording the responses of single photoreceptor cells in vitro; by recording the electroretinogram from the human eye, either with full-field illumination or with spatiotemporal patterns of illumination; and by modelling and theoretical approaches. Part of the research was conducted in Professor Lamb's former laboratory in Cambridge, during the construction of new equipment in the JCSMR.

Multi­focal electro­retinograms

The so-called “multifocal” approach to recording the electroretinogram (mfERG) has been employed to study the spatial distribution of neural responses in the retina. The basic idea is that, if different regions in the visual field are stimulated with temporal sequences of illumination that are uncorrelated with each other, then it becomes possible to extract the signals elicited by stimulation of the individual regions. Of particular interest is the spatial distribution of both sensitivity and circulating current in the photoreceptors (rods and cones). Appropriate equipment and a computer program were obtained through the kind help of Drs A James and T Maddess in the Research School of Biological Sciences, ANU, and the project was undertaken initially with Ms Liz Bowan (PhD student) and subsequently also with Dr Allison Cameron. We have investigated the spatiotemporal pattern of the responses to brief flashes of illumination under photopic (i.e. room-light) conditions. We have shown that the first negative-going wave of the mfERG corresponds closely to the “a-wave” of the conventional full-field photopic ERG, which is known to be elicited by the activity of the cone photoreceptors and the cone OFF bipolar cells. We have examined the spatial profile of these responses, and we have shown that they obey “spatial superposition”; that is, stimulation of a pattern of regions elicits the same response as the sum of responses to stimulation of the individual regions.

Human cone pigment regeneration

Using the more conventional ERG approach, of full-field (or “ganzfeld”) illumination, Omar Mahroo (PhD student, in Cambridge) has been able to measure the time-course of regeneration of cone visual pigment after intense “bleaching” exposures. Previously it has only been possible to measure visual pigment regeneration by an optical technique called reflection densitometry, but we have been able to show that under suitable conditions the regeneration can be extracted using the ERG. Furthermore, we have shown that the time-course of cone pigment regeneration does not follow classical exponential kinetics, but instead the regeneration is rate-limited. We have proposed a model, whereby these kinetics arise from a limitation in the rate at which 11-cis retinal is able to diffuse to opsin in the photoreceptor outer segments.

Inverted responses in single rod photoreceptors

By recording the electrical responses of single rod photoreceptor cells drawn into a suction pipette, Jaakko Järvinen (PhD student, Cambridge) has discovered a previously unreported phenomenon: that light falling on parts of the rod outer segment that are outside the pipette elicit responses of inverted polarity. Previously it had been claimed that there was no response, apart from a small one of normal polarity elicited by scattered light. Voltage-clamp experiments, using simultaneous patch pipette and suction pipette recordings, have shown that the inverted response is induced by the intracellular voltage response (hyperpolarization) to light. And modelling approaches are consistent with the idea that the hyperpolarization triggers an increase in the sodium/calcium exchange activity, which lowers calcium concentration and activates guanylyl cyclase, thereby raising the cGMP level, and opening ion channels, so as to give a response opposite to that of light. Proper consideration of this phenomenon will be important in providing a description of the recovery phase of the light response in photoreceptors.

Molecular mechanisms of rhodopsin's shut-off at the single-photon level

In conjunction with Drs RD Hamer, SC Nicholas, D Tranchina, and PA Liebman, Trevor Lamb undertook a theoretical analysis of the mechanisms involved in shut-off of a single photon-activated rhodopsin molecule. In the past, there has been considerable controversy in the field of phototransduction concerning the molecular mechanism by which it is possible for a single activated molecule to turn off in a reliable and fairly reproducible manner. The results of this study showed that the known process of sequential phosphorylation of activated rhodopsin, via the action of rhodopsin kinase, appears to produce a “ratcheting down” of activity that can account for the experimentally measured reproducibility of single-photon responses.

Human dark adaptation

In collaboration with Professor EN Pugh Jr (Philadelphia, USA), Trevor Lamb undertook an analysis of a wide variety of published results on human dark adaptation and regeneration of visual pigment. The experimental analysis was combined with modelling and mathematical approaches, which showed that the time-course of dark adaptation (and of pigment regeneration) appear to be set by the rate at which opsin can combine with 11‑cis retinal that diffuses down a concentration gradient from the retinal pigment epithelium. One outcome of this research is that it is now possible to obtain an estimate of the relative concentration of the critical retinoid 11‑cis retinal in the living eye through simple non-invasive measurements.

Development of a Scanning Laser Ophthalmoscope for stimulation and densitometry

A collaboration was begun with Dr BA Patterson (Perth), to design and construct a new portable scanning laser ophthalmoscope (SLO), that will not only be able to provide fundus images, but that will also be capable of (a) generating spatial visual stimuli (e.g. for mfERG experiments), and (b) making reflection densitometric measurements of visual pigment from arbitrary regions of the fundus.