Introduction

Parkinson's disease (PD) is one of the most common neurodegenerative diseases¹. The death of dopaminergic neurons in the substantia nigra (SN) and aberrant aggregation of α-synuclein are the core pathological changes of PD^2,3. Iron deposition in the substantia nigra (SN) exacerbates oxidative stress and a-synuclein aggregation, leading to neuronal death in PD^4,5. Our aim was to investigate whether PD disease status modulates the effect of nigral iron deposition on brain function using multimodal magnetic resonance.

Methods

85 PD patients and 140 normal controls (NC) underwent structural T1 magnetic resonance imaging, resting-state functional magnetic resonance imaging and susceptibility weighted imaging scans, as well as motor and non-motor symptoms assessments. According to the median of nigral magnetic susceptibility of NC (0.191ppm), PD and NC were divided into high iron group and normal iron group, respectively. And then ICA method was applied to separate brain large-scale networks.The specific effects of nigral iron deposition on the brain functional networks in PD were investigated by voxel-based interaction analysis.

Results

Mixed effect analysis: Disease* iron interaction effect on brain network function
Based on voxel-based mixed effect analysis, significant interaction effect of disease and nigral iron were observed in several networks: (1) BGN: left inferior frontal gyrus and left insular lobe (peak MNI coordinate: X = -41.5/-33.5, Y = 21.5/3.5, Z = -2.5/13.5; F = 15.8/16.5, respectively) (Figure 1A-B); (2) VN: right middle occipital gyrus, right superior temporal gyrus and bilateral cuneus (peak MNI coordinate: X = 42.5/66.5/4.5, Y = -86.5/-38.5/-82.5, Z = 11.5/13.5/33.5; F = 18.6/22.0/14.4, respectively) (Figure 1C-E).
Post hoc analysis about the FC of PD*iron interaction effect clusters in BGN and VN
Through further post-hoc analysis, we found that the FC of interaction effect brain cluster1 in BGN (left inferior frontal gyrus) in PD-SN_high was higher than that in PD-SN_normal (P = 0.001), while the FC of these brain clusters were not significantly different between NC-SN_high and NC-SN_normal (P = 0.396) (Figure 2A). And the FC of interaction effect brain cluster2 in BGN (left insular lobe) in PD-SN_high was lower than that in PD-SN_normal (P = 0.001), while the FC of these brain clusters was not significantly different between NC-SN_high and NC-SN_normal (P = 0.227) (Figure 2B). Furthermore, the FC of clusters with significant interaction effect in VN was increased in PD-SN_high compared with PD-SN_normal (P = 0.001), while that was decreased in NC-SN_high compared with NC-SN_normal (P < 0.001) (Figure 2C).
Correlations between nigral iron content, network function, and UPDRS II score
Through partial correlation analysis (controlling for gender, age, and education), nigral iron content was significantly positively correlated with UPDRS Ⅱ score in PD (r = 0.366; P = 0.001) (Figure 3D). For BGN, nigral iron content was significantly negatively correlated with the FC of interaction effect brain cluster 2 in BGN (left insular lobe) (r = -0.301; P = 0.006) (Figure 3A), and the FC of interaction effect brain cluster 2 in BGN was negatively correlated with UPDRS Ⅱ score (r = -0.223; P = 0.044) (Figure 3E). For VN, nigral iron content was significantly positively correlated with the FC of interaction effect brain clusters in VN (r = 0.272; P = 0.013) (Figure 3B). Furthermore, there was a significant negative correlation between the FC of interaction effect brain clusters in BGN and VN in PD patients (r = -0.320; P = 0.003) (Figure 3C).
Multiple mediating effect of FC of interaction effect clusters on the relationship between nigral iron and UPDRS Ⅱ score
Multiple mediating analysis revealed that the FC of interaction effect clusters in BGN and mVN partially mediated the association between nigral iron content and UPDRS Ⅱ score (Total indirect effect: Effect = 7.394, Boot SE = 3.875, Boot LLCI = 0.986, Boot ULCI = 16.396). In other words, under separate conditions, neither the FC of interaction effect brain clusters in BGN nor mVN mediated the relationship between nigral iron and UPDRS Ⅱ score (Ind1 indirect effect: Effect = 2.661, Boot SE = 3.526, Boot LLCI = -2.940, Boot ULCI = 11.521; Ind2 indirect effect: Effect = 1.141, Boot SE = 1.012, Boot LLCI = -0.033, Boot ULCI = 4.762, respectively). And the direct effect was significant (Direct effect: Effect = 29.744, Boot SE = 12.542, p = 0.021, Boot LLCI = 5.686, Boot ULCI = 53.893), which indicated that nigral iron deposition could cause disease to worsen directly. Moreover, the FC of interaction effect clusters in BGN and mVN did not serve as chain mediators for the relationship between nigral iron and UPDRS Ⅱ score (Ind3 indirect effect: Effect = 3.592, Boot SE = 3.557, Boot LLCI = -0.335, Boot ULCI = 14.80). This suggested that FC of BGN and mVN play an important role in mediating between nigral iron deposition and disease severity, and neither was indispensable. (Figure 4)

Conclusion

PD disease status specifically moderates the effect of nigral iron deposition on the function of some brain regions in BGN and VN, which partially mediates the relationship between nigral iron deposition and disease severity. Our study provides imaging basis for understanding circuit mechanism of the occurrence and development of PD related to nigral iron deposition.

Acknowledgements

We wish to thank all the participants including patients with Parkinson’s disease and normal volunteers. We also thank the assistance from Department of Neurology in the Second Affiliated Hospital of Zhejiang University School of Medicine.