Francine Carew-Jones

PUBLICATIONS

Coexisting Lewy body disease and clinical parkinsonism in frontotemporal lobar degeneration.

Forrest SL, Crockford DR, Sizemova A, McCann H, Shepherd CE, McGeachie AB, Affleck AJ, Carew-Jones F, Bartley L, Kwok JB, Kim WS, Jary E, Tan RH, McGinley CV, Piguet O, Hodges JR, Kril JJ, Halliday GM

To investigate the prevalence of clinically relevant multiple system atrophy (MSA) and Lewy body disease (LBD) pathologies in a large frontotemporal lobar degeneration (FTLD) cohort to determine if concomitant pathologies underlie the heterogeneity of clinical features. Coexisting LBD in FTLD comprises a small proportion of cases but has implications for clinical and neuropathologic diagnoses and the identification of biomarkers.

Increased Ndfip1 in the substantia nigra of Parkinsonian brains is associated with elevated iron levels.

Howitt J, Gysbers AM, Ayton S, Carew-Jones F, Putz U, Finkelstein DI, Halliday GM, Tan SS

Iron misregulation is a central component in the neuropathology of Parkinson's disease. The iron transport protein DMT1 is known to be increased in Parkinson's brains linking functional transport mechanisms with iron accumulation. The regulation of DMT1 is therefore critical to the management of iron uptake in the disease setting. We previously identified post-translational control of DMT1 levels through a ubiquitin-mediated pathway led by Ndfip1, an adaptor for Nedd4 family of E3 ligases. Here we show that loss of Ndfip1 from mouse dopaminergic neurons resulted in misregulation of DMT1 levels and increased susceptibility to iron induced death. We report that in human Parkinson's brains increased iron concentrations in the substantia nigra are associated with upregulated levels of Ndfip1 in dopaminergic neurons containing α-synuclein deposits. Additionally, Ndfip1 was also found to be misexpressed in astrocytes, a cell type normally devoid of this protein. We suggest that in Parkinson's disease, increased iron levels are associated with increased Ndfip1 expression for the regulation of DMT1, including abnormal Ndfip1 activation in non-neuronal cell types such as astrocytes.

Substantia nigra echomorphology and motor cortex excitability.

Todd G, Taylor JL, Baumann D, Butler JE, Duma SR, Hayes M, Carew-Jones F, Piguet O, Behnke S, Ridding MC, Berg D, Double KL

The aim of our study was to investigate the relation between substantia nigra (SN) echomorphology and indices of motor cortex excitability. Nigral hyperechogenicity in healthy individuals is thought to represent an SN abnormality or predisposition to Parkinson's disease (PD) and its prevalence is greater in the very old. Our study involved 20 old healthy subjects (aged 72-84 years) known to have normal (n=10) or abnormal (n=10) SN echomorphology. All were in good health with no overt neurological signs. SN morphology was assessed with transcranial sonography through the pre-auricular bone window. Motor cortical excitability and intracortical inhibition were assessed with transcranial magnetic stimulation (TMS) over the first dorsal interosseus motor area. Single stimuli were delivered during relaxation and voluntary contraction and paired stimuli were delivered during relaxation. Each cortical hemisphere was analysed separately. The response to single-pulse TMS (in motor cortex ipsilateral to the target SN) did not differ between groups. However, a significant difference between groups was observed in the paired pulse paradigm (conditioning stimulus intensity: 70% resting motor threshold; interstimulus interval: 2 ms). The conditioned motor evoked potential amplitude was significantly larger ipsilateral to the hyperechogenic SN than in controls (P=0.014). Thus, healthy subjects with SN hyperechogenicity exhibit significantly less intracortical inhibition within the motor cortex than subjects with normal echomorphology. Decreased intracortical inhibition is also observed in PD patients. This study provides further evidence that SN hyperechogenicity in healthy individuals is associated with changes characteristic of PD supporting a role for this feature as a vulnerability marker or state marker for subtle nigral dopaminergic dysfunction.

Anti-melanin antibodies are increased in sera in Parkinson's disease.

Double KL, Rowe DB, Carew-Jones FM, Hayes M, Chan DK, Blackie J, Corbett A, Joffe R, Fung VS, Morris J, Riederer P, Gerlach M, Halliday GM

An increasing body of research suggests that a number of immune mechanisms play a role in degenerative pathways in Parkinson's disease (PD). In the current work we investigated a posited humoral immune response in this disorder. Sera from PD patients exhibited a significantly enhanced absorbance response on a novel ELISA for anti-melanin antibodies, compared to sera from age-matched control subjects. The enhanced ELISA absorbance response was specific for catecholamine-based melanins and was unrelated to antiparkinsonian dopaminergic medication. Further, the absorbance response was significantly and negatively correlated with disease duration. These data suggest that a specific humoral anti-melanin antibody response is present in PD and is more active in early disease. While the contribution of this novel immune response to the initiation and progression of this disorder is unclear, this finding supports the hypothesis that specific immune responses occurring in PD may respond to therapeutic interventions in this disorder.

Lipid content determines aggregation of neuromelanin granules in vitro.

Dedov VN, Griffiths FM, Garner B, Halliday GM, Double KL

The neuromelanin pigment of the substantia nigra of the human brain is closely associated with lipids and other non-melanogenic compounds which appear to contribute to the unique and complex morphology of neuromelanin pigment granules. In this work we show that insoluble granules isolated from the human substantia nigra associate in vitro to form pigment aggregates similar to those present in the human brain. Extraction of neuromelanin-associated polar lipids by methanol and/or hexane significantly enhanced melanin aggregate size. A marked (10-fold) increase in granule size was seen after methanol treatment, whereas the application of hexane after methanol reduced this pro-aggregation effect. We have previously reported that hexane and methanol remove the neuromelanin-associated polyisoprenoids dolichol and cholesterol respectively. Thus, the current data suggests that pigment-associated lipids may be a factor regulating pigment aggregation and neuromelanin granule size in vivo.

Alpha-synuclein redistributes to neuromelanin lipid in the substantia nigra early in Parkinson's disease.

Halliday GM, Ophof A, Broe M, Jensen PH, Kettle E, Fedorow H, Cartwright MI, Griffiths FM, Shepherd CE, Double KL

The distribution and tempo of neuronal loss in Parkinson's disease correlates poorly with the characteristic and more widely spread intracellular changes associated with the disease process (Lewy bodies and Lewy neurites). To determine early intracellular changes in regions where cell loss is most marked (dopaminergic A9 substantia nigra) versus regions with Lewy bodies but where cell loss is limited, we assessed 13 patients with definite Parkinson's disease at various disease stages in comparison with controls. Using immunohistochemistry for alpha-synuclein, we confirmed the concentration of this protein in the soma of normal A9 neurons and in Lewy body pathology in brainstem catecholamine neurons in Parkinson's disease. Analysis of the degree of cell loss in brainstem catecholamine cell groups revealed that only the A9 substantia nigra had consistent significant cell loss early in the disease course with greater A9 cell loss correlating with increasing disease duration. To assess the earliest intracellular changes differentiating neurons more likely to degenerate, pigmented A9 and A10 neurons with and without obvious pathology were targeted, cell size and pigment density measured, and intracellular changes in alpha-synuclein location and lipid components analysed at both the light and electron microscope levels. There were no changes observed in healthy A10 neurons in Parkinson's disease compared with controls. Pigmented A9 neurons in later stages of degeneration with obvious Lewy body formation had a significant reduction in intracellular pigment, as previously described. In contrast, A9 neurons of normal morphological appearance and no characteristic pathology in Parkinson's disease exhibited significantly increased pigment density associated with a concentration of alpha-synuclein to the lipid component of the pigment and a loss of associated cholesterol. These changes in vulnerable but apparently healthy A9 neurons occurred without any change in cell size or in the amount of intracellular pigment compared with controls. The increase in pigment density is consistent with previously reported increases associated with oxidation and iron loading, reactions known to precipitate alpha-synuclein. The selectivity of the changes observed in A9 nigral neurons suggests that these early intracellular changes predispose these neurons to more rapid cell loss in Parkinson's disease. The increased concentration of neuronal alpha-synuclein and pigment in normal A9 neurons may already predispose these neurons to precipitate alpha-synuclein around pigment-associated lipid under oxidative conditions. Overall, these changes may trigger a cascade of events leading to larger intracellular aggregates of alpha-synuclein and the dispersement of protective pigment to precipitate cell death in Parkinson's disease.

Differential effects of human neuromelanin and synthetic dopamine melanin on neuronal and glial cells.

Li J, Scheller C, Koutsilieri E, Griffiths F, Beart PM, Mercer LD, Halliday G, Kettle E, Rowe D, Riederer P, Gerlach M, Rodriguez M, Double KL

We investigated the effects of neuromelanin (NM) isolated from the human substantia nigra and synthetic dopamine melanin (DAM) on neuronal and glial cell lines and on primary rat mesencephalic cultures. Lactate dehydrogenase (LDH) activity and lipid peroxidation were significantly increased in SK-N-SH cells by DAM but not by NM. In contrast, iron-saturated NM significantly increased LDH activity in SK-N-SH cells, compared with 100 mg/mL ETDA-treated NM containing a low concentration of bound iron. DAM, but not NM, stimulated hydroxyl radical production and increased SK-N-SH cell death via apoptotic-like mechanisms. Neither DAM nor NM induced any changes in the glial cell line U373. 3H-dopamine uptake in primary rat mesencephalic cultures was significantly reduced in DAM-compared with NM-treated cultures, accompanied by increased cell death via an apoptosis-like mechanism. Interestingly, Fenton-induced cell death was significantly decreased in cultures treated with both Fenton reagent and NM, an effect not seen in cultures treated with Fenton reagent plus DAM. These data are suggestive of a protective role for neuromelanin under conditions of high oxidative load. Our findings provide new evidence for a physiological role for neuromelanin in vivo and highlights the caution with which data based upon model systems should be interpreted.

The dopamine receptor agonist lisuride attenuates iron-mediated dopaminergic neurodegeneration.

Double KL, Halliday GM, Henderson J, Griffiths FM, Heinemann T, Riederer P, Gerlach M

Many dopamine agonists used in the treatment of Parkinson's disease are suggested to be potentially neuroprotective. On the basis of its structure, the dopamine agonist lisuride may share this characteristic. In the current study discrete asymptomatic lesions were produced by the injection of iron-laden neuromelanin into the rat substantia nigra and the animals treated with lisuride to determine the protective potential of this substance. Two treatment regimes were utilised. In the neuroprotective protocol, animals were treated with 0.1 mg.kg(-1) lisuride twice daily 3 days prior to, and 7 days following, the iron lesion. In the neurorescue protocol, the animals received 0.1 mg.kg(-1) lisuride twice daily for 1 week beginning on the fourth day post surgery. Eight weeks post surgery, tyrosine hydroxylase-positive neurons surrounding the injection site (33% of total nigral volume) were counted. Dopamine neuron number in iron-lesioned animals was reduced to 50% of that in vehicle-injected animals. The absence of motoric disturbances or a striatal dopamine deficit in these animals suggests a subclinical dopaminergic lesion. Dopamine neuron number in the quantified area in sham-injected animals receiving lisuride or iron-lesioned animals receiving lisuride in both the neuroprotection and neurorescue groups were not significantly reduced. These results suggest that lisuride can protect neurons against iron-induced cell death and might thus be neuroprotective in Parkinson's disease.