Introduction

Parkinson's disease (PD) and related disorders are characterized by filamentous structures consisting of abnormal α-synuclein in the brains. The propagation of these pathologies is closely correlated with disease progression, and synaptic dysfunction caused by these pathologies is considered to be an early event during the disease progression¹. To develop therapeutic strategies, in vivo detection of the early pathological alterations is critically important. Proton magnetic resonance spectroscopy (MRS) is a non-invasive method that can detect neurotransmitters such as glutamate (Glu). At the resting state under anesthesia, Glu levels measured by MRS (MRS-measured Glu) are known to reflect synaptic densities in the pathological brains². In this work, we aimed to detect early pathological changes caused by accumulation of α-synuclein in vivo using MRS. We hypothesized that MRS-measured Glu can detect an early pathological change, namely synaptic dysfunction, in the α-synuclein accumulation mouse model.

Methods

Mouse Recombinant mouse α-synuclein and fibrils were prepared as described previously³. 12 C57BL/6 wild type male mice were used for this experiment (6 for α-synuclein injection, and 6 for saline injection as control). Under anesthesia with ketamine hydrochloride (71.5 mg/kg, i.p.) and xylazine (14.5 mg/kg, i.p.), the animals were injected with 4 mL of 4 mg/mL mouse α-synuclein fibrils into the bilateral cortex, consecutively. After the MRS measurements described below, the brain was perfused with saline and fixed in 4% PFA in PBS. After cryo-protection in PBS containing 20% sucrose, brains were embedded and frozen in OCT compound, and 20-mm thick sections were prepared by cryostat. Histological analysis: Immunostaining was performed on scanned mice with NeuN and synapsin antibodies for staining of neurons and presynapses, respectively.
MRI and 1H-MRS Both α-synuclein fibril- and saline-injected mice were scanned twice by MRI, both before and 8 weeks after the injections. For MRS measurements, the mice were anesthetized with 1–2% isoflurane and scanned by a 7 T spectrometer (Biospec, AVANCE-III, Bruker Biospin) with a dual-channel phased-array cooled surface coil for transmission and reception (cryoprobe©, Bruker Biospin) using a PRESS sequence (TR/TE = 4000/20 ms). Two volume of interests (VOIs) were localized in the right and left cortex. 256 acquisitions were collected for each cortex. Using total creatine (tCr, the sum of phosphocreatine (PCr) and Cr) signal as a reference, metabolite concentration ratios were calculated using LCModel. In this work, we focused on Glu and total N-acetylaspartate (tNAA, the sum of N-acetylaspartate and N-acetylaspartylglutamate) to test our hypothesis.
Statistical analyses Results were presented as mean ± standard error of the mean. For multiple group comparisons, one-way ANOVA was performed, followed by Fisher’s least significant difference post-hoc test.

Results

Eight weeks after the injection of α-synuclein fibrils, the levels of Glu/tCr were reduced significantly compared to saline-injected mice. There was also significant decrease of tNAA in the α-synuclein fibril-injected mice compared to saline-injected mice. Histological analysis indicated that α-synuclein deposition was propagated from the injected site to the neighboring regions. Synapsin staining demonstrated that presynapses were damaged by α-synuclein accumulation. NeuN staining did not show evidence of neuronal loss, which is in accordance with the currently proposed pathophysiological claim that synaptic dysfunction precedes neurodegeneration¹.

Discussion

In this work, we investigated whether MRS-measured Glu detects synaptic dysfunction caused by pathological α-synuclein accumulation in mouse brain. MRS-measured Glu levels were significantly decreased compared to control mice, suggesting that synaptic dysfunction leads to decreased Glu levels in presynaptic sites. Histological analysis was in agreement with this interpretation, since synapsin protein exists mainly at presynaptic terminals⁴. No evidence of NeuN alterations suggests that neuronal loss has not appeared at this stage. Decreased tNAA levels without evident neuronal loss confirmed by histological analysis may indicate that axonal impairments contributed to the decrease of tNAA, since tNAA is known to exist not only in soma but also in axons5. Mitochondrial dysfunction may also be a potential mechanism of decreased tNAA.

Conclusion

We demonstrated that MRS can detect synaptic dysfunction caused by pathological α-synuclein accumulation in vivo. This method using MRS would be valuable for evaluating therapeutic efficacies of various potential drugs to develop therapeutic strategies for α-synucleinopathies such as Parkinson’s disease and other related disorders.

Acknowledgements

We acknowledge Ms. Shoko Uchida, Mr. Takahiro Shimizu, Mr. Takeharu Minamihisamatsu and other colleagues for valuable help for our experiments.

References

1. Frontiers in neuroscience 2018;12:80. 2. NeuroImage 2014;101:185-192. 3. Acta neuropathologica communications 2014;2:88. 4. The Journal of neuroscience : the official journal of the Society for Neuroscience 2004;24:3711-3720. 5. Progress in neurobiology 2007;81:89-131.