GGC expansions in NOTCH2NLC contribute to Parkinson disease and dopaminergic neuron degeneration

Abstract Background and purpose The role of GGC repeat expansions within NOTCH2NLC in Parkinson's disease (PD) and the substantia nigra (SN) dopaminergic neuron remains unclear. Here, we profile the NOTCH2NLC GGC repeat expansions in a large cohort of patients with PD. We also investigate the role of GGC repeat expansions within NOTCH2NLC in the dopaminergic neurodegeneration of SN. Methods A total of 2,522 patients diagnosed with PD and 1,085 health controls were analyzed for the repeat expansions of NOTCH2NLC by repeat‐primed PCR and GC‐rich PCR assay. Furthermore, the effects of GGC repeat expansions in NOTCH2NLC on dopaminergic neurons were investigated by using recombinant adeno‐associated virus (AAV)‐mediated overexpression of NOTCH2NLC with 98 GGC repeats in the SN of mice by stereotactic injection. Results Four PD pedigrees (4/333, 1.2%) and three sporadic PD patients (3/2189, 0.14%) were identified with pathogenic GGC repeat expansions (larger than 60 GGC repeats) in the NOTCH2NLC gene, while eight PD patients and one healthy control were identified with intermediate GGC repeat expansions ranging from 41 to 60 repeats. No significant difference was observed in the distribution of intermediate NOTCH2NLC GGC repeat expansions between PD cases and controls (Fisher's exact test p‐value = 0.29). Skin biopsy showed P62‐positive intranuclear NOTCH2NLC‐polyGlycine (polyG) inclusions in the skin nerve fibers of patient. Expanded GGC repeats in NOTCH2NLC produced widespread intranuclear and perinuclear polyG inclusions, which led to a severe loss of dopaminergic neurons in the SN. Consistently, polyG inclusions were presented in the SN of EIIa‐NOTCH2NLC‐(GGC)98 transgenic mice and also led to dopaminergic neuron loss in the SN. Conclusions Overall, our findings provide strong evidence that GGC repeat expansions within NOTCH2NLC contribute to the pathogenesis of PD and cause degeneration of nigral dopaminergic neurons.

clear.Here, we profile the NOTCH2NLC GGC repeat expansions in a large cohort of patients with PD.We also investigate the role of GGC repeat expansions within NOTCH2NLC in the dopaminergic neurodegeneration of SN.
Methods: A total of 2,522 patients diagnosed with PD and 1,085 health controls were analyzed for the repeat expansions of NOTCH2NLC by repeat-primed PCR and GC-rich PCR assay.Furthermore, the effects of GGC repeat expansions in NOTCH2NLC on dopaminergic neurons were investigated by using recombinant adeno-associated virus (AAV)-mediated overexpression of NOTCH2NLC with 98 GGC repeats in the SN of mice by stereotactic injection.
Results: Four PD pedigrees (4/333, 1.2%) and three sporadic PD patients (3/2189, 0.14%) were identified with pathogenic GGC repeat expansions (larger than 60 GGC repeats) in the NOTCH2NLC gene, while eight PD patients and one healthy control were identified with intermediate GGC repeat expansions ranging from 41 to 60 repeats.No significant difference was observed in the distribution of intermediate NOTCH2NLC GGC repeat expansions between PD cases and controls (Fisher's exact test p-value = 0.29).
Skin biopsy showed P62-positive intranuclear NOTCH2NLC-polyGlycine (polyG) inclusions in the skin nerve fibers of patient.Expanded GGC repeats in NOTCH2NLC produced widespread intranuclear and perinuclear polyG inclusions, which led to a severe loss of dopaminergic neurons in the SN.Consistently, polyG inclusions were presented in the SN of EIIa-NOTCH2NLC-(GGC)98 transgenic mice and also led to dopaminergic neuron loss in the SN.
Conclusions: Overall, our findings provide strong evidence that GGC repeat expansions within NOTCH2NLC contribute to the pathogenesis of PD and cause degeneration of nigral dopaminergic neurons.

INTRODUC TI ON
As the second most frequent neurodegenerative disorder, Parkinson disease (PD) is characterized by motor symptoms, such as bradykinesia, resting tremor, muscle stiffness, and postural instability, as well as nonmotor symptoms like olfactory dysfunction, sleep disorders, constipation, and dysautonomia.These symptoms are caused by the loss of neurons in various brain regions and may occur before or after the loss of dopaminergic neurons, which are commonly affected in PD [1,2].PD is thought to be caused by a combination of ageing and genetic and environmental risk factors.
To date, more than 20 genes with different degrees of genetic evidence have been found to be mutated in monogenic PD [3][4][5][6].
Although progress has been made, much remains unclear about the genetic etiology of the PD.
In recent years, there has been growing interest in diseases caused by short tandem repeat expansions, which have been associated with various neurodegenerative disorders [7,8].GGC repeat expansions in the GIPC1 locus have been screened as candidate gene for idiopathic PD, as they have been found to be associated with other movement disorders and essential tremor [9][10][11].Previously, our group and others identified GGC repeat expansion in the NOTCH2NLC gene as the genetic cause of neuronal intranuclear inclusion disease (NIID) [12][13][14].Subsequently, NOTCH2NLC with GGC repeat expansions has also been identified in other neurodegenerative disorders [15][16][17].We also reported three different parkinsonism families (3/205) with NOTCH2NLC GGC repeat expansion, indicating that the GGC repeat expansions in NOTCH2NLC could contribute to the pathogenesis of PD [12], which was verified by other group [18].However, the pathogenic GCC repeat expansions within NOTCH2NLC were rare and were not associated with PD in the European population [19][20][21].
Further studies with larger case-control cohorts will help to define the role of GGC repeat expansions within NOTCH2NLC in PD.
In this study, one of the main goals is to establish the prevalence of the NOTCH2NLC GGC repeat expansions in a large number of case-control individuals to further assess whether these repeat expansions are associated with PD.
With the progress in genetic identification that has been made, molecular pathogenesis of NOTCH2NLC GGC repeat expansions is attracting considerable attention [22][23][24][25].However, the pathology of PD patients with the NOTCH2NLC GGC repeat expansions is still poorly understood, and the role of NOTCH2NLC with expanded GGC repeats in substantia nigra (SN) dopaminergic neurons remains unclear [12,17].Here, we examined the effects of GGC repeat expansions within NOTCH2NLC on dopaminergic neurons by expressing NOTCH2NLC with 98 GGC repeats via recombinant adeno-associated viruses (AAVs) in the SN of mice.

Standard protocol approvals, registrations, and patient consents
The study was approved by the ethics committee of Xiangya Hospital Central South University.All participants provided written informed consent.[26].

Participants and clinical assessment
Genomic DNA of peripheral blood leukocytes was obtained from these PD patients, their available family members, and 1085 unaffected healthy control subjects.A comprehensive dataset of basic demographic data, including the subjects' age, gender, family history, disease duration, and clinical features including motor and nonmotor manifestations, was collected from the PD patients enrolled in this study and inputted into the PD-MDCNC.

Animals
The EIIa-NOTCH2NLC-(GGC) 17 and EIIa-NOTCH2NLC-(GGC) 98 transgenic mice, in which NOTCH2NC with normal GGC repeats (17 GGC) or expanded GGC repeats (98 GGC) was expressed ubiquitously, were established in our previous study [22].The transgenic mice and wild-type C57BL/6J mice were bred and maintained on a 12:12-h light/dark cycle (lights off at 7 p.m.).Two-month-old wildtype male mice were used for stereotaxic injection.All animal procedures were performed in accordance with the institutional guidelines of the Animal Care and Use Committee at Central South University.

Constructs
The 5′-untranslated region (5′UTR) of transcript variant 1 of NOTCH2NLC with 98 GGC repeats was cloned into pX551 vector using the following primers: 5UTR_F: ATA GGT ACC ACC GGT GCT GAG GCG GCG GCC GAG GAG CG and 5UTR_R: ATA GGA TCC CAC AGG GTT CAT AGC CAT CTC GAC ACT GCA ATG CAT GCG CGG GGG TCG CGC A. A 3*HA tag in reading frame with polyglycine (polyG) was then fused to the C-terminal of 5′UTR to obtain a NOTCH2NLC-(GGC) 98 -HA construct.

Stereotaxic injection of AAVs into mouse brain
Stereotaxic surgery was performed as described previously [4].

Cell culture and transfection
Neuro-2a cells were cultured at 37°C with 5% CO 2 in Dulbecco modified Eagle medium supplemented with 10% fetal bovine serum and 100 U/mL of penicillin/streptomycin. Plasmid transient transfection on Neuro-2a cells was performed using Lipofectamine 2000 reagent following the manufacturer's instruction.

Immunofluorescent staining
Mice were anesthetized and perfused intracardially with 0.9% NaCl, followed by 4% paraformaldehyde.Isolated mouse brains were dehydrated in 30% sucrose and then sectioned at 30 μm.
Brain sections were blocked in 3% bovine serum albumin in 0.3% Triton X-100/phosphate-buffered saline for 1 h followed by incubation with primary antibodies at 4°C overnight.After incubated with Alexa Fluor-conjugated secondary antibodies and 4,6-diamidino-2-phenylindole, the brain sections were mounted and visualized with a Zeiss scope.

Quantification and statistical analysis
All statistical analyses were performed using SPSS 26.0 and R version 4.1.1.All quantification data were presented as mean ± SEM.
Fisher exact test was used for genetic analysis.Linear regression and Pearson correlation were used to calculate the relationship between GGC repeat sizes and age at onset of PD, and the correlation coefficient r was determined.A p-value of <0.05 was considered statistically significant.Among the participants, 333 cases had a family history of the disease, of whom 205 had been described and analyzed in our previous publication.

Identification of NOTCH2NLC GGC repeat expansions in patients with PD
We identified four PD pedigrees (4/333, 1.2%) and three sporadic PD patients (3/2189, 0.14%) carrying pathogenic GGC repeat expansions (≥60 GGC repeats) in the NOTCH2NLC gene.The GGC repeat expansion sizes ranged from 66 to 156, with an average of 91.71 ± 27.02 GGC repeats.Notably, we found GGA interruptions at the 3′ end of the GGC repeat expansions were the primary forms of interruptions in these patients with NOTCH2NLC GGC expansion.
Furthermore, eight patients with PD were identified to carry intermediate GGC repeat expansions ranging from 41 to 60 repeats in the NOTCH2NLC gene, whereas only one healthy control participant was found to have an intermediate GGC repeat expansions (41 repeats) in this gene.The carrier rate of intermediate GGC repeat expansions in NOTCH2NLC among PD patients was slightly higher than that in healthy individuals (0.31% vs. 0.09%, respectively).Despite this, there was no significant difference in the distribution of intermediate NOTCH2NLC GGC repeat expansions between PD cases and controls (Fisher exact test p = 0.29, odds ratio = 3.459, 95% confidence interval = 0.432-27.690; Figure 1).

Clinical features of PD patients harboring GGC repeat expansions in NOTCH2NLC gene
A total of 13 patients with PD were found to have pathogenic GGC repeat expansions in the NOTCH2NLC gene, and eight patients with PD were found to carry intermediate GGC repeat expansions.Those PD patients harboring GGC repeat expansions in NOTCH2NLC gene have been confirmed with no pathogenic or likely pathogenic mutations in the established causative genes of PD by whole-exome sequencing or whole-genome sequencing and multiplex ligation-dependent probe amplification in our previous description [5,6].The clinical characteristics of these patients are summarized in Table S1.3).Furthermore, we examined the histopathological features in the skin biopsy of PD patients.NOTCH2NLC with GGC repeat expansion has been reported to produce NOTCH2NLC-polyG inclusions in NIID patients.However, it remains unknown whether the same histopathological feature exists in PD patients with this mutation.Double-immunofluorescent staining identified intranuclear polyG inclusions in the skin of case PD-11 with 90 GGC repeat expansions by using the anti-PEP122 antibody that was prepared and validated in our previous study [22].These polyG inclusions were found colocalized with P62, a typical pathological marker in NIID (Figure 4a).Meanwhile, we also detected phosphorylated α-synuclein in dermal nerve fibers in PD patients but did not find α-synuclein deposition (Figure 4b).

Overexpression of NOTCH2NLC-(GGC) 98 causes dopaminergic neuron degeneration in the SN in an age-dependent manner
Although our genetic findings suggested that GGC repeat expansions in NOTCH2NLC were potentially associated with PD, whether these expansions can cause degeneration of dopaminergic neurons in the SN, a key pathological hallmark of PD, remains unknown.To test this possibility, we made a mutant NOTCH2NLC construct.An unconventional AUG codon residing in the 5′UTR upstream of the GGC repeats has been verified to produce a short polyG-containing protein named uN2CpolyG, which has been proved to be the major protein of NOTCH2NLC [22][23][24].We thus cloned the 5′UTR of the

Immunofluorescent staining of HA and GFP verified their expression
in the injected brain regions (Figure 5e).Interestingly, no significant loss of TH + neurons was found in the ventral tegmental area (Figure 6b).

F I G U R E 2
Correlation between age at onset and the size of NOTCH2NLC GGC repeat.Linear regression and Pearson correlation were used to calculate the association between GGC repeat sizes and age at onset of PD, and the correlation coefficient r was determined.The x-axis gives the age at onset, and the y-axis gives the number of GGC repeats.Expansion (+): patients with NOTCH2NLC GGC repeat expansions.Expansion (−): patients without NOTCH2NLC GGC repeat expansions.
Because dopaminergic neurons in the SN project processes to the striatum, we then measured the fluorescence intensity of TH immunostaining in the striatum.As expected, the TH signal in the caudate-putamen (CPu) was partially lost, and TH fibers to the nucleus accumbens (NAc) and olfactory tubercle (OT) were indistinguishable from the control side 1 month after NOTCH2NLC-(GGC) 98 -HA infection (Figure 6c,d).After 2 months, the TH signal in the CPu was almost completely lost, whereas it was partially preserved in the NAc and OT (Figure 6c,d).Collectively, these findings suggest that the expression of NOTCH2NLC with expanded GGC repeats may trigger dopaminergic neuron degeneration in an age-dependent manner in mice.
Because AAV viral delivery had the potential to induce cellular stress and toxicity to enhance polyG toxicity, we then detected the expression of polyG and the survival of TH + neurons in the SN of EIIa-NOTCH2NLC-(GGC) 17 and EIIa-NOTCH2NLC-(GGC) 98 transgenic mice, which were established in our previous study [22].As expected, intranuclear or perinuclear polyG inclusions were identified in the TH + neurons of 2-month-old EIIa-NOTCH2NLC-(GGC) 98 transgenic mice (Figure 6e).Notably, the number of TH + neurons was reduced significantly in EIIa-NOTCH2NLC-(GGC) 98 transgenic mice when compared to the EIIa-NOTCH2NLC-(GGC) 17 transgenic mice (Figure 6f).

DISCUSS ION
In this study, we conducted a case-control analysis to investigate the association between GGC repeat expansions in the NOTCH2NLC gene and PD.Our findings revealed that among a large PD cohort, four PD pedigrees and three sporadic PD patients had pathogenic GGC repeat expansions (>60 repeats) in the NOTCH2NLC gene.Our results confirmed that expanded GGC repeats in the NOTCH2NLC gene were associated with PD [12,18].Furthermore, we performed a longitudinal follow-up study on 13 carriers with pathogenic NOTCH2NLC GGC repeat expansions.Those patients were confirmed with the diagnosis of PD, and they showed typical PD symptoms such as good response to levodopa and dopaminergic neuronal loss detected by dopamine transporter PET.To gain further insights, we will conduct a longitudinal clinicopathological investigation on PD patients carrying these repeat expansions.
In addition, eight PD patients and one healthy control were identified with intermediate-length NOTCH2NLC GGC repeat expansions (41-60 repeats), but the association between the intermediate-length NOTCH2NLC GGC repeat expansions and PD is not well established in our study.It had been reported that intermediate-length GGC repeat expansions in NOTCH2NLC may be associated with PD in two different studies, but such intermediate repeat is consistent with a primary risk factor for neurodegenerative diseases such as Alzheimer disease, PD, and other repeat-expansion-mediated diseases.Given that neurons are terminally differentiated cells that do not divide, aging-related factors, including greater vulnerability and reduced protein homeostasis, make the neurons more susceptible to damage than glia cells.According to current reports, the potential mechanism of NOTCH2NLC-polyGmediated toxicity in our mouse model may be due to DNA repair alterations or nucleocytoplasmic transport disruption, which was explored in the cell and mouse models of NIID, another neurodegenerative disorder caused by the same mutation in NOTCH2NLC [22][23][24][25].In addition, like other misfolded proteins, accumulated NOTCH2NLC-polyG protein aggregates may recruit, through protein-protein interactions, other proteins such as RNA-binding proteins and transcription factors, thereby sequestering them from their functional locations and affecting their activities.
Future studies are necessary to further investigate the molecular pathogenesis of PD.
We also examined the phosphorylated alpha-synuclein in SN but did not find positive signals in the mice injected with AAV-NOTCH2NLC-(GGC) 98 .Our data may indicate that polyG inclusions did not accelerate phosphorylated alpha-synuclein deposition at least 2 months after injection of AAV-NOTCH2NLC-(GGC) 98 .However, GGC repeat expansion of NOTCH2NLC may impact alpha-synuclein accumulation or clearance over time.It is well known that repeat expansion disorders such as Huntington disease and spinocerebellar ataxia types 1-3 have been found to be associated with dysfunction of the autophagy system, and they often involve abnormal folding and aggregation of proteins, resulting in disturbance of protein homeostasis within the cell.Thus, it is possible that the clearance of other proteins such as alpha-synuclein will be affected, and the phenotype may become more significant over time.However, further research is required to explore the impact of GGC repeat expansions in NOTCH2NLC and alpha-synuclein.
The GGC repeat expansions of NOTCH2NLC is found in many diseases, including neuronal intranuclear inclusion disease, essential tremor, PD, Alzheimer disease, multiple system atrophy, and even oculopharyngodistal myopathy.How could the GGC repeat expansion of NOTCH2NLC result in different disease phenotypes?It seems that GGC repeat size and the types of trinucleotide interruption may lead to different phenotypes of disease.Generally, the repeat size of muscle weakness-dominant phenotype is largest, and parkinsonism-dominant phenotype is smallest.Dementia-dominant and essential tremor-dominant phenotypes usually have a purer GGC repeat.There are also other possible mechanisms, including the toxicity of polyG-containing protein, the toxicity of repeat RNA, and the methylation status of NOTCH2NLC [17,22].
In conclusion, our study provides genetic and functional evidence to support the role of GGC repeat expansions in NOTCH2NLC in the pathogenesis of PD.

F I G U R E 1
Figure 2).All patients with NOTCH2NLC GGC repeat expansions were followed up for 3-13 years and were eventually diagnosed with clinically established PD or probable PD according to the Movement Disorder Society clinical diagnostic criteria for PD.Brain magnetic resonance imaging showed white matter hyperintensities in some patients, but cerebellar atrophy was not observed in any patients.11 C-2β-carbomethoxy-3β-(4-fluorophenyl) tropane positron emission tomography (PET) imaging from case PD-11 revealed a graded and asymmetrical reduction in dopamine transporter binding in the putamen, which was compatible with PD (Figure 3).

NOTCH2NLC
gene harboring 98 GGC repeats from an NIID patient and fused it with a 3*HA tag to facilitate detection of the polyGcontaining protein (Figure 5a,b).We first checked the expression of NOTCH2NLC-(GGC) 98 -HA and GFP control constructs by transfecting them in cultured HEK293 cells.Immunofluorescent staining showed abundant NOTCH2NLC-polyG aggregates in cells expressing NOTCH2NLC-(GGC) 98 -HA construct, whereas cells expressing GFP control construct exhibited homogeneous cytoplasmic GFP fluorescence (Figure 5c).Then, the NOTCH2NLC-(GGC) 98 -HA and GFP control constructs were packaged into AAV9 viruses and bilaterally injected into the SN of adult wide-type C57BL/6 mice upon stereotaxic injection (Figure 5d).
Next, we evaluated the effects of GGC repeat expansions within NOTCH2NLC on dopaminergic neurons in vivo.One month after injection, expression of NOTCH2NLC-(GGC) 98 -HA led to widespread formation of intranuclear or perinuclear polyG inclusions in the dopaminergic neurons labeled with TH.Importantly, TH-positive (TH + ) dopaminergic neurons showed reduced fluorescence in the axons and dendrites and had loose distribution in the NOTCH2NLC-(GGC) 98 -HA-injected side compared to the GFP-injected side (Figure6a).Additionally, the number of TH + neurons in the SN was significantly reduced by 26% compared to the GFP-injected side 1 month after injection of NOTCH2NLC-(GGC) 98 -HA.The reduction further increased to 69% 2 months after injection (Figure6b).