Substantia Nigra Ferric Overload and Neuromelanin Loss in Parkinson's
Total Page:16
File Type:pdf, Size:1020Kb
medRxiv preprint doi: https://doi.org/10.1101/2021.04.13.21255416; this version posted April 20, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license . Iron and neuromelanin in PD using 7T MRI - Page | 1 Substantia nigra ferric overload and neuromelanin loss in Parkinson’s disease measured with 7T MRI Catarina Ruaa,b, Claire O’Callaghanc,d, Rong Yeb, Frank H. Hezemansb,f, Luca Passamontib,e, P Simon Jonesb, Guy B Williamsa, Christopher T Rodgersa, James B Roweb,f Affiliations: a) Wolfson Brain Imaging Centre, Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK b) Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK c) Department of Psychiatry, University of Cambridge, UK d) Brain and Mind Centre and School of Medical Sciences, Faculty of Medicine and Health, University of Sydney, Australia e) Istituto di Bioimmagini e Fisiologia Molecolare (IBFM), Consiglio Nazionale delle Ricerche (CNR), Milano, Italia f) MRC Cognition and Brain Sciences Unit, University of Cambridge, Cambridge, UK Corresponding author: Name: Catarina Rua Address: The Wolfson Brain Imaging Centre, Box 65, Cambridge Biomedical Campus, Cambridge, CB2 0QQ, UK Telephone: +44 7709 774 939 e-mail address: [email protected] Keywords: 7T MRI; Parkinson’s disease; Substantia Nigra; Iron; Neuromelanin. NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 1 medRxiv preprint doi: https://doi.org/10.1101/2021.04.13.21255416; this version posted April 20, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license . Iron and neuromelanin in PD using 7T MRI - Page | 2 1 Abstract: 2 3 Background: Vulnerability of the substantia nigra dopaminergic neurons in Parkinson’s 4 disease is associated with ferric overload, leading to neurodegeneration with cognitive and 5 motor decline. Here, we quantify iron and neuromelanin-related markers in vivo using ultra- 6 high field 7-Tesla MRI, and examine the clinical correlates of these imaging assessments. 7 8 Methods: Twenty-five people with mild-to-moderate Parkinson’s disease and twenty-six 9 healthy controls underwent high-resolution imaging at 7-Tesla with a T2*-weighted sequence 10 (measuring susceptibility-χ and R2*, sensitive to iron) and a magnetization transfer-weighted 11 sequence (MT-w, sensitive to neuromelanin). From an independent control group (N=29), we 12 created study-specific regions-of-interest for five neuromelanin- and/or iron-rich subregions 13 within the substantia nigra. Mean R2*, susceptibility-χ and their ratio, as well as the MT-w 14 contrast-to-noise ratio (MT-CNR) were extracted from these regions and compared between 15 groups. We then tested the relationships between these imaging metrics and clinical severity. 16 17 Results: People with Parkinson’s disease showed a significant ~50% reduction in MT-CNR 18 compared to healthy controls. They also showed a 1.2-fold increase in ferric iron loading ∗ 19 (elevation of the ∆�!/∆� ratio from 0.19±0.058ms/ppm to 0.22±0.059ms/ppm) in an area of 20 the substantia nigra identified as having both high neuromelanin and susceptibility MRI signal 21 in healthy controls. In this region, the ferric-to-ferrous iron loading was associated with 22 disease duration (b=0.0072, pFDR=0.048) and cognitive impairment (b=-0.0115, pFDR=0.048). 23 24 Conclusions: T2*-weighted and MT-weighted high-resolution 7T imaging markers identified 25 neurochemical consequences of Parkinson’s disease, in overlapping but not-identical regions. 26 These changes correlated with non-motor symptoms. 27 28 29 30 2 medRxiv preprint doi: https://doi.org/10.1101/2021.04.13.21255416; this version posted April 20, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license . Iron and neuromelanin in PD using 7T MRI - Page | 3 31 1. Introduction 32 33 Iron accumulation and loss of pigmented neuromelanin cells in the substantia nigra are key 34 parts of the complex neuropathology of Parkinson’s disease (1–6). While iron is essential for 35 normal brain homeostasis, it can become toxic by the generation of free radicals promoted 36 by redox reactions, converting ferrous Fe2+ to the ferric Fe3+ form (7). In Parkinson’s disease, 37 iron overload and increased oxygen free radical formation induce lipid peroxidation of cell 38 membranes (8), promoting dopaminergic neurodegeneration (9). In contrast, neuromelanin 39 can provide a neuroprotective pathway by chelating iron and hence preventing damage by 40 Fe3+ (10,11). However, in Parkinson’s disease, the buffering capacity of neuromelanin may be 41 exceeded, leading to neurotoxicity, oxidative stress, neuroinflammation, and ultimately loss 42 of neurons (10,12). 43 These pathological features in the substantia nigra of patients with Parkinson’s disease have 44 been extensively confirmed in post mortem studies (1,2,4,6,13). However, characterising 45 these changes and distinguishing different types of iron in vivo remains a challenge. Recently, 46 specialised MRI sequences have been developed to characterise tissue iron and changes in 47 neuromelanin content. With sufficient resolution and sensitivity, these could potentially be 48 used to identify functionally distinct sub-regions of the substantia nigra. The ventral portion 49 of the substantia nigra, known as the substantia nigra pars reticulata (SNr), has high 50 concentration of iron in healthy subjects. The dorsal portion, known as the substantia nigra 51 pars compacta (SNpc), is rich in neuromelanin, the pigmented by-product of dopamine and 52 noradrenaline metabolism (11,14). 53 The dopaminergic-rich SNpc can be imaged in vivo using a neuromelanin-sensitive 54 “magnetisation-transfer weighted” (MT-w) MRI sequence (15–17). The MT-w signal intensity 55 correlates with the location and density of the neuromelanin content in older adults (18). 56 Alternative sequences have been used to observe changes in iron in the healthy brain and in 57 disease (19,20) by measuring the T2* transverse relaxation time (or its inverse rate, R2*) or 58 the local tissue magnetic susceptibly χ using quantitative susceptibility mapping (QSM) (21– 59 23). Susceptibility weighted imaging (SWI) uses post-processing of T2*-w images, with the 60 magnitude and filtered phase information providing image contrast from iron storage (24). 61 SWI has been shown to provide unique contrast to identify the appearance of the nigrossome- 3 medRxiv preprint doi: https://doi.org/10.1101/2021.04.13.21255416; this version posted April 20, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license . Iron and neuromelanin in PD using 7T MRI - Page | 4 62 1 (N1-sign) in healthy controls, and its loss in patients with Parkinson’s disease (24). 63 Susceptibility imaging is sensitive to any molecule that generates an MR phase shift. In 64 contrast to SWI, both QSM and R2* provide quantitative measures of susceptibility effects. 65 However, they are unable on their own to differentiate the sources of susceptibility, and 66 therefore an adjunctive method is required. One approach is to capitalise on the fact that 67 ferrous and ferric irons show differences in relaxivity per unit concentration (25), which can 68 be estimated from the R2*-to-χ ratio. We therefore take the R2*-to-χ ratio as a proxy of ferric 69 vs. ferrous loading, and use this ratio as a biomarker for iron accumulation in Parkinson’s 70 disease. 71 In this study, we used high resolution dedicated MRI sequences for in vivo imaging of 72 neuromelanin with high resolution (0.08 mm3) and iron (0.34 mm3) using ultra-high field 7T, 73 based on magnetisation transfer (26) and multi-echo T2*, respectively, with nigral regions of 74 interest derived from an independent cohort of age-matched healthy controls (27). The 75 contrast-to-noise ratio (CNR) of the MT signal was interpreted as a measure of neuromelanin 76 content, whereas χ and R2* were used to quantify iron in the tissue. We investigated (1) the 77 molecular changes of neuromelanin and iron in sub-regions of the substantia nigra in 78 Parkinson’s disease compared to healthy controls, and (2) the association between imaging 79 markers and cognitive and motor functions in the patient group. We hypothesised that people 80 with Parkinson’s disease would show decreased MT signal contrast and increased χ and R2* 81 in the SNpc. Since ferric iron demonstrates higher susceptibility and R2* values than ferrous 82 iron at the same concentration in tissue (25,28) we also expected an increase of the R2*-to- χ 83 ratio in the SNpc. 84 2. Methods 85 86 2.1 Participants 87 88 Twenty-six healthy controls (HC, 15 male, age 65±5 years) and twenty-five participants with 89 Parkinson’s disease (PD, 18 male, age 67±7 years) were recruited via the University of 90 Cambridge Parkinson’s disease research clinic and the Parkinson’s UK volunteer network. The 91 study was approved by the Cambridge Research Ethics Committee (16/EE/0084; 92 10/H0308/34) and all participants provided written informed consent in accordance with the 4 medRxiv preprint doi: https://doi.org/10.1101/2021.04.13.21255416; this version posted April 20, 2021.