Neuromelanin Magnetic Resonance Imaging of the Substantia Nigra in Huntington’s Disease

Abstract

Background:

Huntington’s disease (HD) is an autosomal-dominant neurodegenerative disorder inducing motor, psychiatric changes and cognitive decline, characterized pathologically by striatal atrophy. Pathological changes in the extra-striatal structures, such as the substantia nigra (SN), and abnormalities in pre-synaptic striatal dopamine neurotransmission are also known to occur. Neuromelanin (NM)-sensitive magnetic resonance imaging (NM-MRI) is an innovative technique that was recently developed allowing the in vivo study of pathological changes in the dopaminergic neurons of the SN.

Objective:

To investigate the SN MR signal in HD patients.

Methods:

We performed a cross-sectional study using a specific T1-weighted MR sequence to visualize NM. The areas and signal intensity contrast ratios of the T1 hyperintense SN regions were obtained using a semi-automatic segmentation method.

Results:

A total of 8 HD patients and 12 healthy subjects were evaluated. The SN area was markedly reduced in the HD group compared with the control group (p = 0.02), even after normalization of the SN area with the midbrain area and age correction (p = 0.01). There was a significant reduction in the intensity contrast ratio of the hyperintense SN areas to crus cerebri in HD patients comparing with controls (p = 0.04) after correction for age.

Conclusions:

NM-sensitive MR techniques were used for the first time to study the SN in HD patients, showing loss of NM in this region, supporting the implication of dopaminergic neuronal changes in disease pathology. Future research needs to be conducted to evaluate the potential of SN area and intensity contrast as biomarkers for HD.

Keywords

Huntington’s disease neuromelanin-sensitive magnetic resonance imaging neurodegeneration substantia nigra

INTRODUCTION

Huntington’s disease (HD) is an autosomal dominant inherited neurodegenerative disorder, caused by a Cytosine-Adenine-Guanine (CAG) expansion in the HTT (huntingtin) gene that codes for the huntingtin protein [1], clinically characterized by a triad of gradually worsening motor, cognitive and psychiatric symptoms, with chorea being its most prominent feature [2]. The neuropathological hallmark of HD is progressive bilateral degeneration of the striatum due to the loss of medium spiny neurons (MSNs), which starts up to 15 years prior to the onset of motor symptoms [3]. It is still uncertain if this neurodegeneration process begins in the striatum with retrograde degeneration of the cortico-striatal pathway, or if striatal afferents, such as the nigrostriatal dopaminergic pathway, have a primary role in promoting striatal degeneration [4]. Several studies have reported nigrostriatal dopaminergic neuron degeneration and loss of substantia nigra (SN) neurons in HD [5 –8]. Also, an in vitro study has shown evidence of mutant huntingtin interfering with the transcription of the enzyme tyrosine hydroxylase, a rate-limiting enzyme for dopamine biosynthesis, whose activity is decreased in HD [8].

Dopaminergic neurons of the SN, particularly in the SN pars compacta, contain neuromelanin (NM), a highly paramagnetic dark pigment thought to be formed by oxidative polymerization of dopamine or noradrenaline, with a possible neuroprotective role in brain oxidative damage by free radicals and ferric iron [9,10, 9,10]. However, NM has also been linked to neurodegeneration, through disruption of iron homeostasis [9]. Hence, striatal volume loss, associated with the dysregulation of the nigrostriatal pathway and the possible disruption of iron homeostasis in HD [11], may be linked to SN degeneration and NM depletion.

Few imaging studies have investigated the nigrostriatal pathway in HD. Only recently have specific T1-weighted MRI sequences been developed to detect in vivo SN NM signal changes [12]. These have been widely used to quantitatively investigate SN degeneration in Parkinson’s disease, but, to the best of our knowledge, had not previously been used to study HD.

The aim of this proof-of-principle study was to test whether SN signal changes can be found in HD, using a specific MRI sequence to visualize NM (NM-MRI). We hypothesize that this sequence may be useful in the future to further investigate the role of NM in HD.

MATERIALS AND METHODS

Participants

We performed a cross-sectional study comparing HD patients with healthy individuals with no signs or family history of a neurodegenerative disorder. HD patients were recruited from the Movement Disorders Outpatient Clinic of the Hospital de Santa Maria in Lisbon, Portugal. All patients were scored for motor (TMS) and functional capacity (TFC) according to the Unified Huntington’s Disease Rating Scale (UHDRS). Healthy control (HC) subjects were recruited from hospital staff and relatives. 20 subjects were included in the study: 8 HD patients with abnormal HTT CAG repeats expansion (≥40) and 12 HC. One HD patient had a Westphal HD variant (67/20 HTT CAG repeats). Subjects’ clinical characteristics are summarized in Table 1.

Table 1

Sample demographic and clinical characteristics. Data are presented as median [range]

	HC (n = 12)	HD (n = 8)	p-value^*
Gender (M/F)	7 : 5	3 : 5	0.40
Age (years)	62 [49, 83]	46 [27, 81]	0.02
Disease duration (years)	NA	7.5 [1, 16]	NA
UHDRS-TMS	NA	20 [4, 71]	NA
UHDRS-TFC	NA	12 [1, 13]	NA
CAG length	NA	44 [40, 67]^†	NA

^*Mann-Whitney U-test. ^†Molecular tests confirmed HD in all patients, but the exact number of repeats was available only for 7 of the 8 patients; HD, Huntington’s disease; TFC, total functional capacity; TMS, total motor score; UHDRS, unified Huntington’s disease rating scale; CAG, cytosine-adenine-guanine triplet; NA, not applicable. Statistically significant results are in bold.

All participants were examined by a movement disorders specialist. Participants with contraindications to perform an MR study were excluded. Patients with severe motor symptoms were not recruited due to the need for subjects to remain still during the long acquisition times required for NM-MRI (approximately 8 minutes). All examinations were performed with the written consent from each subject, with approval from the local ethics committee, and in compliance with national legislation and the Declaration of Helsinki guidelines.

Imaging protocol

MRI data was acquired using a 3.0 Tesla Philips scanner (Philips Achieva; Philips Medical Systems, Best, Netherlands). A T1 Fast Spin Echo NM-sensitive sequence, with parameters similar to previous publications by our group was used [13]. Additionally, whole-brain three-dimensional T1-weighted images and FLAIR axial images were acquired and evaluated to exclude other pathological findings that could interfere with NM assessments.

Imaging analysis

Analysis of NM-MRI images was performed using OsiriX^® software (v. Lite 9.5; Pixmeo, Geneva, Switzerland) by one of the authors, blinded to all clinical data and genetic status.

For each subject, the slice containing the largest area of high-signal in the SN was selected for analysis. Semi-automatic region growing segmentation was performed as previously described in [14], starting with the selection of a seed point in the anterior region of the SN and using a region-growing algorithm to append neighboring pixels. Since midbrain atrophy is expected in HD patients [3], it is important to consider inter-subject anatomical variability when comparing the SN areas between patients and controls. With this in mind, the midbrain was manually segmented to obtain a global midbrain area, and the ratio of SN (SNarea) to midbrain area (Marea), hereafter termed SN ratio, was calculated and used for further comparison of SN areas. The contrast-to-background ratio (CBR) of each hyperintense SN area was also determined, by placing a reference region in the crus cerebri (Fig. 1).

Fig. 1

Method for calculating the CBR. An 8 mm² circular region (yellow) was placed in the crus cerebri (CC) on each side. The CBR was calculated by dividing the SNc mean intensity (green) by the mean intensity in the corresponding CC.

Measurements of the SN were performed on both left and right sides, and because no significant differences were found between sides, the average value was used. Additionally, we corrected NM results for age according to the results from Xing et al. (2018) [15], which measured the physiological life span pigmentation changes of the SN by NM-MRI. More specifically, the expected SN CBR values and SN ratios for each individual at a certain age were calculated respectively from the nonlinear fitting of the CBR versus age and from the linear fitting of the normalized SN volumes versus age obtained in Xing et al. (2018) [15] and subtracted from our data.

Statistical analysis

All statistical analyses were performed using the R software (v.3.5.1; R Foundation for Statistical Computing, Vienna, Austria). Non-parametric tests were used. The Wilcoxon signed rank test for paired samples was used to assess side-wise differences in parameters measured in each brain hemisphere. Differences between HD patients and HC were evaluated with the Mann-Whitney U test. A p-value lower than 0.05 was considered significant.

RESULTS

No significant differences were found in gender between HD and HC subjects (p = 0.40) but HC were significantly older than HD patients (p = 0.02) motivating correction for age effects as previously described. HD patients had a median disease duration of 7.5 years (range: 1–16 years).

The acquired NM-MR images allowed a clear identification of the SN hyperintense area in all subjects, which were therefore included in the analysis.

Using semi-automatic segmentation of the SN hyperintense area, a median (range) SN area of 25 (17–36) mm² was obtained for the HD group and an area of 35 (22–50) mm², was obtained for the HC group (Table 2; Fig. 2). The measured SN high-signal area was markedly reduced in the HD group compared with the control group (p = 0.02).

Table 2

SN NM measurements in HD and HC. Data presented as median [range]

	HC	HD	p-value^*
SN area (mm²)	35 [22,50]	25 [17,36]	0.02
Midbrain area (cm²)	6.6 [5.6,7.6]	5.7 [5.0,7.5]	0.03
SN ratio^a (%)	5.2 [3.5,7.4]	4.4 [3.1–5.1]	0.03
SN ratio^b (%)	1.6 [0.4,4.4]	–0.8 [–1.8,2.9]	0.01
CBR^a (%)	14 [11,17]	13 [8,16]	0.57
CBR^b (%)	2.3 [–7.6,19]	–4.8 [–10,13]	0.04

^*Mann-Whitney U-test. HC, healthy control group; HD, Huntington’s disease group; SN, substantia nigra; CBR- contrast-to-background ratio; ^abefore and ^bafter correction for age.

Fig. 2

Area of the SN high-intensity region (mm²) on neuromelanin-sensitive MR images in patients with HD and HC compared using a Mann-Whitney U-test. The label ** indicates statistically significant differences. SN, substantia nigra; HD, Huntington’s disease; HC, healthy controls.

A significant reduction of midbrain area was also found in the HD group (p = 0.03), with a median (range) value of 5.7 (5.0–7.5) cm² in HD patients, compared with a midbrain area of 6.6 (5.6–7.6) cm² in the HC group (Table 2). After normalization of the SN area with the midbrain area, the SN ratio measured in the HD group was 4.4 (3.1–5.1)% and was significantly reduced, with a p-value of 0.03, when compared with the ratio of 5.2 (3.5–7.4)% measured for the HC group. After correcting for age, the difference in SN ratios was even more pronounced (p = 0.01) as shown in Fig. 3.

Fig. 3

SN Ratio (a) before and (b) after correcting SN area for age (b), compared using Mann-Whitney U-tests. The label ** indicates statistically significant differences. SN, substantia nigra; HD, Huntington’s disease; HC, healthy controls.

The median (range) CBR of the SN hyperintensity in the HD group were 13 (8–16)%, which were not significantly different from those measured for the control group: 14 (11–17)% (Table 2; Fig. 4), with the CBR values of HD patients being only slightly reduced when compared to controls (p = 0.57). However, after correcting for age effects, the CBR values were significantly reduced in HD patients compared with controls (p = 0.04).

Fig. 4

Contrast-to-background (CBR) ratio (%) of the SN versus the crus cerebri (a) before and (b) after correction for age (c) compared using a Mann-Whitney U-test. The labels NS./* indicate statistically non-significant/significant differences. SN, substantia nigra; HD, Huntington’s disease; HC, healthy controls.

For the patient with Westphal HD variant an SN area of 18.7 mm² was measured, corresponding to an SN ratio of 3.7%. The CBR obtained for the same patient was 11%. Although the TMS and TFC for this patient were the highest and lowest in the studied sample, corresponding to scores of 71 and 1, respectively, the measured area and SN ratio were not the lowest observed. In any case, both were close to the bottom range of values observed, and the CBR was within the lowest quartile.

DISCUSSION

This is the first study using NM-MRI on HD, showing a reduction of NM in the SN of HD patients in vivo. The observed reduction of the SN area in HD patients suggests that in HD there is a depletion of NM in the SN. Since signal intensity in the SN in NM-MR images has been shown to be closely related to the quantity of NM-containing neurons [16], the reduced CBR found in HD favors the notion that the quantity of pigmented neurons in HD is decreased compared to HC. Indeed, an in vitro study [5] has revealed a decreased cross-sectional area of the SN, a reduction in the number of neurons in the SN and a depletion in the amount of intraneuronal melanin in the SN of HD patients.

In previous MRI studies, an atrophy of the SN in HD was described using different techniques, such as voxel based morphometric [17], diffusion MRI [18], and T1 inversion-recovery sequences [7]. The innovation in our approach is the use of an MRI technique that is specific to study NM and that does not seem to be significantly influenced by paramagnetic iron effects [19]. As NM is present mainly in dopaminergic neurons of the SN pars compacta, a reduction of the SN hyperintense area in these images suggests loss of these neurons. The SN area is related to the spatial distribution of the NM within the SN. On the other hand, the CBR is sensitive to the signal intensity, and it has been shown to reflect NM-concentration [16]. The information conveyed by these metrics is hence not exactly the same and therefore a perfect correlation between the two would not be expected (see Supplementary Material).

Our study has some limitations that need to be considered, mainly the small number of HD patients included, which impeded further data analysis, such as analysis of the association between NM areas and clinical features or disease duration (scatter plots are provided as Supplementary Material for illustrative purposes only). Moreover, the HD group was heterogeneous in terms of disease duration and disease stage. Furthermore, one of the patients had a juvenile Westphal variant. Since NM has consistently been shown to be reduced in patients with Parkinson’s disease, where rigidity and bradykinesia are key features, we hypothesize that NM reduction may be more prominent in HD patients with higher rigidity/bradykinesia. Still, although we expected the highest neuromelanin imaging changes in the Westphal variant patient, presenting a severe hypokinetic-rigid Westphal phenotype, a patient with the classical phenotype had an even higher SN neuromelanin area reduction on NM-MRI (SN area of 17 mm², SN ratio 3.1%). Considering that only one patient with this type of variant could be studied, these results should be interpreted with caution, and further investigations are warranted.

Additionally, as the HC group was significantly older (all controls were over 47 years old, and hence within the range of ages for which NM shows a decrease with age [15]), this might have been a confounding factor in this study. To mitigate this problem, the measurements were corrected for age as previously described.

To conclude, our data supports the occurrence of SN neural loss in symptomatic HD patients, particularly dopaminergic SN pars compacta neurons that send afferent projections to the striatum. Still, whether this is a trigger or a consequence of striatal degeneration remains uncertain. Longitudinal studies of NM are required to assess the time course of SN degeneration in HD patients and its relation to basal ganglia atrophy.

The potential value of this imaging technique as a biomarker is currently unknown and was not evaluated as part of this study. Future research is required in order to validate the sensitivity and reliability of this imaging technique as a marker of disease progression.

CONFLICTS OF INTEREST

The authors declare that there are no conflicts of interest relevant to this work.

Footnotes

ACKNOWLEDGMENTS

We thank all the patients and control subjects for their time and commitment to this research. Fundação para a Ciência e a Tecnologia for financial support through grants FCT – IF/00364/2013, UID/EEA/50009/2013 (RL and RGN).

The supplementary material is available in the electronic version of this article: .

References

MacDonald

, Ambrose

, Duyao

, Myers

, Lin

, Srinidhi

, et al.A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell1993;72:971–83. doi: 10.1016/0092-8674(93)90585-E.

Bates

, Dorsey

, Gusella

, Hayden

, Kay

, Leavitt

, et al.Huntington disease. Nat Rev Dis Prim2015;1:15005. doi: 10.1038/nrdp.2015.5.

Vonsattel

, Myers

, Stevens

, Ferrante

, Bird

, Richardson

. Neuropathological classification of Huntington’s disease. J Neuropathol Exp Neurol1985;44:559–77.

Rosas

, Salat

, Lee

, Zaleta

, Pappu

, Fischl

, et al.Cerebral cortex and the clinical expression of Huntington’s disease: Complexity and heterogeneity. Brain2008;131:1057–68. doi: 10.1093/brain/awn025.

Oyanagi

, Takeda

, Takahashi

, Ohama

, Ikuta

. A quantitative investigation of the substantia nigra in Huntington’s disease. Ann Neurol1989;26:13–9. doi: 10.1002/ana.410260103.

Bamford

, Robinson

, Palmiter

, Joyce

, Moore

, Meshul

. Dopamine modulates release from corticostriatal terminals. J Neurosci2004;24:9541–52. doi: 10.1523/JNEUROSCI.2891-04.2004.

Kiferle

, Mazzucchi

, Unti

, Pesaresi

, Fabbri

, Nicoletti

, et al.Nigral involvement and nigrostriatal dysfunction in Huntington’s disease: Evidences from an MRI and SPECT study. Park Relat Disord2013;19:800–5. doi: 10.1016/j.parkreldis.2013.05.005.

Yohrling

IV GJ

, Jiang

GCT

, DeJohn

, Miller

, Young

, Vrana

, et al.Analysis of cellular, transgenic and human models of Huntington’s disease reveals tyrosine hydroxylase alterations and substantia nigra neuropathology. Mol Brain Res2003;119:28–36. doi: 10.1016/j.molbrainres.2003.08.009.

Zucca

, Basso

, Cupaioli

, Ferrari

, Sulzer

, Casella

, et al.Neuromelanin of the human substantia Nigra: An update. Neurotox Res2014;25:13–23. doi: 10.1007/s12640-013-9435-y.

10.

Sulzer

, Zecca

. Intraneuronal dopamine-quinone synthesis: A review. Neurotox Res1999;1:181–95. doi: 10.1007/bf03033289.

11.

Muller

, Leavitt

. Iron dysregulation in Huntington’s disease. J Neurochem2014;130:328–50. doi: 10.1111/jnc.12739.

12.

Sasaki

, Shibata

, Tohyama

, Takahashi

, Otsuka

, Tsuchiya

, et al.Neuromelanin magnetic resonance imaging of locus ceruleus and substantia nigra in Parkinson’s disease. Neuroreport2006;17:1215–8. doi: 10.1097/01.wnr.0000227984.84927.a7.

13.

Reimão

, Pita Lobo

, Neutel

, Guedes

, Coelho

, Rosa

, et al.Substantia nigra neuromelanin-MR imaging differentiates essential tremor from Parkinson’s disease. Mov Disord2015;30:953–9. doi: 10.1002/mds.26182.

14.

Correia Guedes

, Reimão

, Paulino

, Nunes

, Bouça-Machado

, Abreu

, et al.Neuromelanin magnetic resonance imaging of the substantia nigra in LRRK2 -related Parkinson’s disease. Mov Disord2017;32:1331–3. doi: 10.1002/mds.27083.

15.

Xing

, Sapuan

, Dineen

, Auer

. Life span pigmentation changes of the substantia nigra detected by neuromelanin-sensitive MRI. Mov Disord2018;33(11):1792–9. doi: 10.1002/mds.27502.

16.

Kitao

, Matsusue

, Fujii

, Miyoshi

, Kaminou

, Kato

, et al.Correlation between pathology and neuromelanin MR imaging in Parkinson’s disease and dementia with Lewy bodies. Neuroradiology2013;55:947–53. doi: 10.1007/s00234-013-1199-9.

17.

Douaud

, Gaura

, Ribeiro

, Lethimonnier

, Maroy

, Verny

, et al.Distribution of grey matter atrophy in Huntington’s disease patients: A combined ROI-based and voxel-based morphometric study. Neuroimage2006;32:1562–75. doi: 10.1016/j.neuroimage.2006.05.057.

18.

Douaud

, Behrens

, Poupon

, Cointepas

, Jbabdi

, Gaura

, et al.In vivo evidence for the selective subcortical degeneration in Huntington’s disease. Neuroimage2009;46:958–66. doi: 10.1016/j.neuroimage.2009.03.044.

19.

Reimão

, Ferreira

, Nunes

, Pita Lobo

, Neutel

, Abreu

, et al.Magnetic resonance correlation of iron content with neuromelanin in the substantia nigra of early-stage Parkinson’s disease. Eur J Neurol2016;23:368–74. doi: 10.1111/ene.12838.