J B Furness

University of Melbourne, Melbourne, Victoria, Australia

Are you J B Furness?

Claim your profile

Publications (396)1225.45 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Disorders of gastrointestinal functions that are controlled by enteric neurons commonly accompany fatty liver disease. Established fatty liver disease is associated with diabetes, which itself induces enteric neuron damage. Here, we investigate the relationship between fatty liver disease and enteric neuropathy, in animals fed a high-fat, high-cholesterol diet in the absence of diabetes.
    Neurogastroenterology and Motility 06/2014; · 2.94 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Discovery of adequate pharmacological treatments for constipation has proven elusive. Increased numbers of bowel movements were reported as a side-effect of ulimorelin treatment of gastroparesis, but there has been no investigation of the site of action. Anesthetized rats were used to investigate sites and mechanisms of action of ulimorelin. Intravenous ulimorelin (1-5 mg/kg) caused a substantial and prolonged (~1 h) increase in colorectal propulsive activity and expulsion of colonic contents. This was prevented by cutting the nerves emerging from the lumbosacral cord, by the nicotinic receptor antagonist hexamethonium and by antagonists of the ghrelin receptor. The effect of intravenous ulimorelin was mimicked by direct application of ulimorelin (5 μg) to the lumbosacral spinal cord. Ulimorelin is a potent prokinetic that causes propulsive contractions of the colorectum by activating ghrelin receptors of the lumbosacral defecation centers. Its effects are long-lasting, in contrast with other colokinetics that target ghrelin receptors.
    Neurogastroenterology and Motility 11/2013; · 2.94 Impact Factor
  • J B Furness, D P Poole
    [Show abstract] [Hide abstract]
    ABSTRACT: The functioning of the gastrointestinal tract is under the control of the most extensive system of peripheral neurons in the body, the enteric nervous system, and the largest endocrine system of the body, the GEP endocrine system. The enteric nervous system in large mammals contains 500 million neurons, and the GEP endocrine system produces more than 30 hormones. Numerous enteric neuropathies affecting both humans and animals have been described and digestive disorders affect commercially important species, such as horses and cattle. The most severe enteric neuropathies (e.g., lethal white syndrome in horses or Hirschsprung's disease in humans) can be fatal. Also, horses with ileus or other digestive disorders are commonly euthanized. In this review we discuss examples of enteric neuropathies that affect agricultural animals and humans: prion disease, postoperative ileus, distal enteric aganglionosis, and infective diarrhea. Enteric neurons and glia are a location of prion proteins and are involved in transmission of the infection from gut to brain and brain to gut. Postoperative ileus is a complex disorder involving the local inhibitory effects of sympathetic nervous system activation and the release of opioids, presumably from enteric neurons. Intestinal inflammation, especially of the external muscle that includes enteric ganglia, also occurs in ileus. Congenital distal bowel aganglionosis, responsible for lethal white syndrome in horses, Hirschsprung's disease in humans, and similar conditions in mice and rats, is a fatal condition if untreated. Mutations of the same genes can cause the condition in each of these species. The only effective current treatment is surgical removal of the aganglionic bowel. Infectious diarrheas involve activation of enteric secretomotor neurons by pathogens and the toxins they produce, which causes substantial fluid loss. Strategies to target enteric neurons in the treatment of secretory diarrheas have not been developed. Disorders of enteroendocrine cells, other than GEP endocrine tumors, are less well documented. However, evidence for the involvement of gut endocrine cells in a subset of patients with irritable bowel syndrome, and in the symptomology of celiac disease, has been demonstrated. Further investigation of the involvement of enteric neural and endocrine signaling systems in digestive disorders, especially in agricultural and companion animals, may lead to diagnostic and therapeutic advances.
    Journal of Animal Science 12/2011; 90(4):1203-12. · 2.09 Impact Factor
  • Source
    L R Rivera, D P Poole, M Thacker, J B Furness
    [Show abstract] [Hide abstract]
    ABSTRACT: Nitric oxide (NO), produced by the neural nitric oxide synthase enzyme (nNOS) is a transmitter of inhibitory neurons supplying the muscle of the gastrointestinal tract. Transmission from these neurons is necessary for sphincter relaxation that allows the passage of gut contents, and also for relaxation of muscle during propulsive activity in the colon. There are deficiencies of transmission from NOS neurons to the lower esophageal sphincter in esophageal achalasia, to the pyloric sphincter in hypertrophic pyloric stenosis and to the internal anal sphincter in colonic achalasia. Deficits in NOS neurons are observed in two disorders in which colonic propulsion fails, Hirschsprung's disease and Chagas' disease. In addition, damage to NOS neurons occurs when there is stress to cells, in diabetes, resulting in gastroparesis, and following ischemia and reperfusion. A number of factors may contribute to the propensity of NOS neurons to be involved in enteric neuropathies. One of these is the failure of the neurons to maintain Ca(2+) homeostasis. In neurons in general, stress can increase cytoplasmic Ca(2+), causing a Ca(2+) toxicity. NOS neurons face the additional problem that NOS is activated by Ca(2+). This is hypothesized to produce an excess of NO, whose free radical properties can cause cell damage, which is exacerbated by peroxynitrite formed when NO reacts with oxygen free radicals.
    Neurogastroenterology and Motility 09/2011; 23(11):980-8. · 2.94 Impact Factor
  • Autonomic Neuroscience 09/2011; 163(s 1–2):58–59. · 1.85 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Ghrelin and ghrelin receptor agonist have effects on central neurons in many locations, including the hypothalamus, caudal brain stem, and spinal cord. However, descriptions of the distributions of ghrelin-like immunoreactivity in the CNS in published work are inconsistent. We have used three well-characterized anti-ghrelin antibodies, an antibody to the unacylated form of ghrelin, and a ghrelin peptide assay in rats, mice, ghrelin knockout mice, and ghrelin receptor reporter mice to re-evaluate ghrelin presence in the rodent CNS. The stomach served as a positive control. All antibodies were effective in revealing gastric endocrine cells. However, no specific staining could be found in the brain or spinal cord. Concentrations of antibody 10 to 30 times those effective in the stomach bound to nerve cells in rat and mouse brain, but this binding was not reduced by absorbing concentrations of ghrelin peptide, or by use of ghrelin gene knockout mice. Concentrations of ghrelin-like peptide, detected by enzyme-linked immunosorbent assay in extracts of hypothalamus, were 1% of gastric concentrations. Ghrelin receptor-expressing neurons had no adjacent ghrelin immunoreactive terminals. It is concluded that there are insignificant amounts of authentic ghrelin in neurons in the mouse or rat CNS and that ghrelin receptor-expressing neurons do not receive synaptic inputs from ghrelin-immunoreactive nerve terminals in these species.
    Neuroscience 07/2011; 193:1-9. · 3.12 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Animal proof of principle study. To determine whether capromorelin, a compound that causes defecation by stimulating ghrelin receptors within the lumbosacral defecation centers, is effective after spinal cord injury (SCI), and whether SCI significantly alters sensitivity to the compound. University of Melbourne and Austin Hospital, Melbourne, Australia. Rats were subjected to spinal cord contusion injury or were sham-operated. At 6 weeks after surgery, effects of capromorelin on blood pressure, heart rate and propulsive contractions of the colorectum were investigated. Capromorelin caused robust propulsive activity in the colorectum soon after its application. The compound was similarly effective in naïve, sham-operated and spinal cord-injured rats. Blood pressure increases caused by capromorelin were not exaggerated after SCI, and there was no evidence of phasic blood pressure increases when the colon was contracted by the compound. Capromorelin is a therapeutic compound that could potentially be used to relieve constipation by triggering defecation in spinal cord-injured patients.
    Spinal Cord 05/2011; 49(10):1036-41. · 1.90 Impact Factor
  • Source
    M Thacker, L R Rivera, H-J Cho, J B Furness
    [Show abstract] [Hide abstract]
    ABSTRACT: Damage to mucosal epithelial cells, muscle cells and enteric neurons has been extensively studied following intestinal ischemia and reperfusion (I/R). Interestingly, the effects of intestinal I/R on enteric glia remains unexplored, despite knowledge that glia contribute to neuronal maintenance. Here, we describe structural damage to enteric glia and associated changes in distribution and immunoreactivity of the neuronal protein Hu. The mouse small intestine was made ischemic for 3 h and reperfused from 1 to 12 h. Immunohistochemical localisation of glial fibrillary acidic protein (GFAP), Hu and TUNEL were used to evaluate changes. At all time points glial cells became distorted, which was evident by their altered GFAP immunoreactivity, including an unusual appearance of bright perinuclear GFAP staining and the presence of GFAP globules. The numbers of neurons per ganglion area were significantly fewer in ganglia that contained distorted glia when compared with ganglia that contained glia of normal appearance. The distribution of Hu immunoreactivity was altered at all reperfusion time points. The presence of vacuoles and Hu granules in neurons was evident and an increase in nuclear Hu, relative to cytoplasmic Hu, was observed in ganglia that contained both normal and distorted glial cells. A number of neurons appeared to lose their Hu immunoreactivity, most noticeably in ganglia that contained distorted glial cells. TUNEL reaction occurred in a minority of glial cells and neurons. Structural damage to gliofilaments occurs following I/R and may be associated with damage to neighboring neurons.
    Neurogastroenterology and Motility 03/2011; 23(11):e500-9. · 2.94 Impact Factor
  • Autonomic Neuroscience-basic & Clinical - AUTON NEUROSCI-BASIC CLIN. 01/2011; 163(1):58-58.
  • Autonomic Neuroscience-basic & Clinical - AUTON NEUROSCI-BASIC CLIN. 01/2011; 163(1):69-69.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The intrinsic primary afferent neurons (IPANs) in the intestine are the first neurons of intrinsic reflexes. Action potential currents of IPANs flow partly through calcium channels, which could feasibly be targeted by pregabalin. The aim was to determine whether pregabalin-sensitive α2δ1 subunits associate with calcium channels of IPANs and whether α2δ1 subunit ligands influence IPAN neuronal properties. We used intracellular electrophysiological recording and in situ hybridisation to investigate calcium channel subunit expression in guinea-pig enteric neurons. Key The α subunits of N (α1B) and R (α1E) type calcium channels, and the auxiliary α2δ1 subunit, were expressed by IPANs. This is the first discovery of the α2δ1 subunit in enteric neurons; we therefore investigated its functional role, by determining effects of the α2δ1 subunit ligand, pregabalin, that inhibits currents carried by channels incorporating this subunit. Pregabalin (10 μmol L(-1)) reduced the action potential duration. The effect was not increased with increase in concentration to 100 μmol L(-1). If N channels were first blocked by ω-conotoxin GVIA (0.5 μmol L(-1)), pregabalin had no effect on the residual inward calcium current. Reduction of the calcium current by pregabalin substantially inhibited the after-hyperpolarising potential (AHP) and increased neuron excitability. Intrinsic primary afferent neurons express functional N (α1B) channel-forming subunits that are associated with α2δ1 modulatory subunits and are inhibited by pregabalin, plus functional R (α1E) channels that are not sensitive to binding of pregabalin to α2δ subunits. The positive effects of pregabalin in irritable bowel syndrome (IBS) patients might be partly mediated by its effect on enteric neurons.
    Neurogastroenterology and Motility 10/2010; 22(10):e301-8. · 2.94 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The distribution and chemical phenotypes of sympathetic and dorsal root ganglion (DRG) neurons innervating the equine ileocecal junction (ICJ) were studied by combining retrograde tracing and immunohistochemistry. Immunoreactivity (IR) for tyrosine hydroxylase (TH), dopamine beta-hydroxylase (DBH), neuronal nitric oxide synthase (nNOS), calcitonin gene-related peptide (CGRP), substance P (SP), and neuropeptide Y (NPY) was investigated. Sympathetic neurons projecting to the ICJ were distributed within the celiac (CG), cranial mesenteric (CranMG), and caudal mesenteric (CaudMG) ganglia, as well as in the last ganglia of the thoracic sympathetic chain and in the splanchnic ganglia. In the CG and CranMG 91 +/- 8% and 93 +/- 12% of the neurons innervating the ICJ expressed TH- and DBH-IR, respectively. In the CaudMG 90 +/- 15% and 94 +/- 5% of ICJ innervating neurons were TH- and DBH-IR, respectively. Sympathetic (TH-IR) fibers innervated the myenteric and submucosal ganglia, ileal blood vessels, and the muscle layers. They were more concentrated at the ICJ level and were also seen encircling myenteric plexus (MP) and submucosal plexus (SMP) descending neurons that were retrogradely labeled from the ICJ. Among the few retrogradely labeled DRG neurons, nNOS-, CGRP-, and SP-IR nerve cells were observed. Dense networks of CGRP-, nNOS-, and SP-IR varicosities were seen around retrogradely labeled prevertebral ganglia neurons. The CGRP-IR fibers are probably the endings of neurons projecting from the intestine to the prevertebral ganglia. These findings indicate that this crucial region of the intestinal tract is strongly influenced by the sympathetic system and that sensory information of visceral origin influences the sympathetic control of the ICJ.
    The Journal of Comparative Neurology 10/2010; 518(19):4046-66. · 3.66 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We have previously demonstrated that a centrally penetrant ghrelin receptor agonist enhances colorectal motility, through activation of the lumbo-sacral defecation center (L6-S1 region of the spinal cord) in rats. In the present study, we examined the effects of the native peptide and its non-acylated counterpart in eliciting this stimulatory effect on colorectal motility. Rats were anesthetised with α-chloralose and ketamine, and colorectal intraluminal pressure and propelled intraluminal liquid volume were recorded in vivo. Intrathecal application of acylated ghrelin to the L6-S1 region of the spinal cord, but not intravenous application, elicited groups of phasic increases in colorectal intraluminal pressure that were associated with increased fluid output through the anal cannula. The effect was dose-dependent. The colokinetic effects of ghrelin were prevented if the pelvic nerves were severed. Reverse transcription polymerase chain reaction revealed the expression of the ghrelin and ghrelin receptor genes in the lumbo-sacral spinal cord. In contrast to acylated ghrelin, des-acyl ghrelin failed to cause changes in colorectal motility. However, when des-acyl ghrelin and ghrelin were applied simultaneously at the L6-S1 region, the ghrelin-induced enhancement of colorectal motility was significantly attenuated. It is concluded that acylation of the ghrelin peptide is essential to promote propulsive contractions of the colorectum and that des-acyl ghrelin opposes this effect. At most other sites of ghrelin action, des-acyl ghrelin either has no effect or it mimics ghrelin. This is the first evidence that non-acylated ghrelin opposes the action of the acylated peptide in the spinal cord.
    Neurogastroenterology and Motility 10/2010; 22(10):1124-31. · 2.94 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: As they migrate through the developing gut, a sub-population of enteric neural crest-derived cells (ENCCs) begins to differentiate into neurons. The early appearance of neurons raises the possibility that electrical activity and neurotransmitter release could influence the migration or differentiation of ENNCs. The appearance of neuronal sub-types in the gut of embryonic mice was examined using immunohistochemistry. The effects of blocking various forms of neural activity on ENCC migration and neuronal differentiation were examined using explants of cultured embryonic gut. Nerve fibers were present in close apposition to many ENCCs. Commencing at E11.5, neuronal nitric oxide synthase (nNOS), calbindin and IK(Ca) channel immunoreactivities were shown by sub-populations of enteric neurons. In cultured explants of embryonic gut, tetrodotoxin (TTX, an inhibitor of action potential generation), nitro-L-arginine (NOLA, an inhibitor of nitric oxide synthesis) and clotrimazole (an IK(Ca) channel blocker) did not affect the rate of ENCC migration, but tetanus toxin (an inhibitor of SNARE-mediated vesicle fusion) significantly impaired ENCC migration as previously reported. In explants of E11.5 and E12.5 hindgut grown in the presence of TTX or tetanus toxin there was a decrease in the number nNOS+ neurons close to the migratory wavefront, but no significant difference in the proportion of all ENCC that expressed the pan-neuronal marker, Hu. (i) Some enteric neuron sub-types are present very early during the development of the enteric nervous system. (ii) The rate of differentiation of some sub-types of enteric neurons appears to be influenced by TTX- and tetanus toxin-sensitive mechanisms.
    Neurogastroenterology and Motility 05/2010; 22(5):e127-37. · 2.94 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Agonists of ghrelin receptors can lower or elevate blood pressure, and it has been suggested that the increases in blood pressure are caused by actions at receptors in the spinal cord. However, this has not been adequately investigated, and the locations of neurons in the spinal cord that express ghrelin receptors, through which blood pressure increases may be exerted, are not known. We investigated the effects within the spinal cord of two non-peptide ghrelin receptor agonists, GSK894490 and CP464709, and two peptide receptor agonists, ghrelin and des-acyl ghrelin, and we used polymerase chain reaction (PCR) and in situ hybridization to examine ghrelin receptor expression. I.v. application of the non-peptide ghrelin receptor agonists caused biphasic changes in blood pressure, a brief drop followed by a blood pressure increase that lasted several minutes. The blood pressure rise, but not the fall, was antagonized by i.v. hexamethonium. Application of these agonists or ghrelin peptide directly to the spinal cord caused only a blood pressure increase. Des-acyl ghrelin had no significant action. The maximum pressor effects of agonists occurred with application at spinal cord levels T9 to T12. Neither i.v. nor spinal cord application of the agonists had significant effect on heart rate or the electrocardiogram. Ghrelin receptor gene expression was detected by PCR and in situ hybridization. In situ hybridization localized expression to neurons, including autonomic preganglionic neurons of the intermediolateral cell columns at all levels from T3 to S2. The numbers of ghrelin receptor expressing neurons in the intermediolateral cell columns were similar to the numbers of nitric oxide synthase positive neurons, but there was little overlap between these two populations. We conclude that activation of excitatory ghrelin receptors on sympathetic preganglionic neurons increases blood pressure, and that decreases in blood pressure caused by ghrelin agonists are mediated through receptors on blood vessels.
    Neuroscience 03/2010; 166(2):671-9. · 3.12 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: RS-127445 is a selective, high affinity 5-HT(2B)receptor antagonist. We investigated whether 5-HT(2B)receptor antagonists can reduce colonic visceral hypersensitivity caused by restraint stress or by proximal colonic inflammation. Visceral hypersensitivity was induced in rats by either restraint stress or injection of 2, 4, 6-trinitrobenzene sulfonic acid (TNBS) into the proximal colon. Restraint stress produced a significant increase in numbers of abdominal contractions evoked by colorectal distension (CRD), measured as a quantitative index of visceral nociception in rats. Seven days after TNBS injection, the pain threshold to CRD at the non-inflamed distal colon, that was determined as the minimum pressure required to evoke abdominal cramp, was significantly decreased. The effect of RS-127445 on visceral hypersensitivity was assessed in either naïve or TNBS-treated rats. Oral administration of a selective, high affinity 5-HT(2B)receptor antagonist, RS-127445, significantly inhibited visceral hypersensitivity provoked by restraint stress (35 to 74% inhibition at 1 to 10 mg kg(-1)). Oral RS-127445 produced a significant suppression of TNBS-induced visceral hypersensitivity (15 to 62% inhibition at 3 to 30 mg kg(-1)), although it was without significant effect on the visceral nociceptive threshold of naïve rats. RS-127445 (1 to 30 mg kg(-1), p.o.) also dose-dependently reduced the restraint stress-induced defecation in naïve and TNBS-treated rats. These results suggest that 5-HT(2B)receptors are involved in signaling from the colon in rats in which there is visceral hypersensitivity and that a selective 5-HT(2B)receptor antagonist could have therapeutic potential for the treatment of gut disorders characterized by visceral hypersensitivity.
    Neurogastroenterology and Motility 09/2009; 22(2):e69-76. · 2.94 Impact Factor
  • Source
    A E Lomax, K A Sharkey, J B Furness
    [Show abstract] [Hide abstract]
    ABSTRACT: Knowledge of neural circuits, neurotransmitters and receptors involved in the sympathetic regulation of gastrointestinal (GI) function is well established. However, it is only recently that the interaction of sympathetic neurons, and of sympathetic transmitters, with the GI immune system and with gut flora has begun to be explored. Changes in the behaviour of sympathetic nerves when gut function is compromised, for example in ileus and in inflammation, have been observed, but the roles of the sympathetic innervation in these and other pathologies are not adequately understood. In this article, we first review the principal roles of the sympathetic innervation of the GI tract in controlling motility, fluid exchange and gut blood flow in healthy individuals. We then discuss the evidence that there are important interactions of sympathetic transmitters with the gut immune system and enteric glia, and evidence that inflammation has substantial effects on sympathetic neurons. These reciprocal interactions contribute to pathological changes in ways that are not yet clarified. Finally, we focus on inflammation, diabetes and postoperative ileus as conditions in which there is sympathetic involvement in compromised gut function.
    Neurogastroenterology and Motility 09/2009; 22(1):7-18. · 2.94 Impact Factor
  • K. Nurgali, M. Thacker, L. Pontell, J. B. Furness
    Autonomic Neuroscience 08/2009; 149(1):97-98. · 1.85 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are active at resting membrane potential and thus contribute to neuronal excitability. Their increased activity has recently been demonstrated in models of nerve injury-induced pain. The major aim of the current study was to investigate altered HCN channel protein expression in trigeminal sensory neurons following inflammation of the dura. HCN1 and HCN2 channel immunoreactivity was observed on the membranes of medium- to large-sized trigeminal ganglion neurons with 76% and 85% of HCN1 and HCN2 expressing neurons also containing the 200 kDa neurofilament protein (associated with myelinated fibers). Western immunoblots of lysates from rat trigeminal ganglia also showed bands with appropriate molecular weights for HCN1 and HCN2. Three days after application of complete Freund's adjuvant (CFA) to the dura mater, Western blot band densities were significantly increased; compared to control, to 166% for HCN1 and 284% for HCN2 channel protein. The band densities were normalized against alpha-actin. In addition, the number of retrogradely labeled neurons from the dura expressing HCN1 and HCN2 was significantly increased to 247% (HCN1) and 171% (HCN2), three days after inflammation. When the opioid receptor partial agonist, buprenorphine, was given systemically, immediately after CFA, the inflammation-induced increase in HCN protein expression in both Western blot and immunohistochemical experiments was not observed. These results suggest that HCN1 and HCN2 are involved in inflammation-induced sensory neuron hyperexcitability, and indicate that an opioid receptor agonist can reverse the protein upregulation.
    Neuroscience 06/2009; 162(2):453-61. · 3.12 Impact Factor
  • Autonomic Neuroscience-basic & Clinical - AUTON NEUROSCI-BASIC CLIN. 01/2009; 149(1):97-97.

Publication Stats

14k Citations
1,225.45 Total Impact Points

Institutions

  • 1969–2014
    • University of Melbourne
      • • Department of Anatomy and Neuroscience
      • • Department of Physiology
      • • Department of Zoology
      Melbourne, Victoria, Australia
  • 1970–2011
    • Victoria University Melbourne
      Melbourne, Victoria, Australia
  • 2008–2010
    • Gifu University
      • Department of Basic Veterinary Science
      Gihu, Gifu, Japan
  • 2009
    • Queen's University
      • Department of Physiology
      Kingston, Ontario, Canada
  • 2007
    • Federal University of Minas Gerais
      Cidade de Minas, Minas Gerais, Brazil
  • 2005
    • University of Bologna
      • School of Agriculture and Veterinary Medicine
      Bologna, Emilia-Romagna, Italy
  • 2001
    • Murdoch Childrens Research Institute
      Melbourne, Victoria, Australia
  • 1993
    • Deakin University
      • Department of Biological Sciences
      Geelong, Victoria, Australia
  • 1990–1993
    • Flinders Medical Centre
      Tarndarnya, South Australia, Australia
    • Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo
      San Paulo, São Paulo, Brazil
  • 1978–1993
    • Flinders University
      • • School of Medicine
      • • Centre for Neuroscience
      • • Department of Anatomy and Histology
      Adelaide, South Australia, Australia
  • 1991
    • The Royal Children's Hospital
      Melbourne, Victoria, Australia
  • 1986
    • Neuroscience Research Australia
      Sydney, New South Wales, Australia
  • 1980
    • National Institute of Mental Health (NIMH)
      • Laboratory of Clinical Science
      Maryland, United States
  • 1977
    • Prince Henry's Institute
      Melbourne, Victoria, Australia
  • 1972
    • University of Birmingham
      Birmingham, England, United Kingdom