An imbalance of netrin‐1 and DCC during nigral degeneration in experimental models and patients with Parkinson's disease

Abstract Aims Multiple guidance cues, such as netrin‐1 (NTN‐1)/deleted in colorectal carcinoma (DCC), control the guidance of axons and help establish functional neural circuits during development. However, the function of these guidance molecules during the neurodegenerative process is unclear. Methods To access the alterations of NTN‐1 and DCC during the onset and progression of PD, we first established two subacute and one chronic PD model. Then, we investigated the relationship between the NTN‐1/DCC pathway and cell death in SH‐SY5Y cells. Finally, we conducted correlation studies between plasma NTN‐1 and parkinsonian symptoms in patients to understand how this pathway contributes to PD. Results We found that the imbalance of NTN‐1 and DCC was a common feature of nigral DA neuron injury in PD mouse models. We investigated that MPP+ inhibited NTN‐1 expression and increased DCC expression in a concentration‐ and time‐dependent manner. We further discovered a significant decrease in plasma NTN‐1 levels and a positive correlation with UPDRS scores in PD patients. Conclusion Our findings confirmed the imbalance of NTN‐1/DCC signaling during nigral degeneration in experimental PD models and found for the first time a correlation of plasma NTN‐1 with PD symptoms in patients.


| INTRODUC TI ON
Human brains contain hundreds of billions of nerve cells that form a complex network of connections and provide the substrate for information processing. Depending on the environment and experience, this neural network can be constantly reconnected and engineered.
The imbalances in network signals can cause toxicity, damage, and death of neurons, which then impairs neural communication processes, leading to neurological diseases. [1][2][3][4][5] Growing evidence suggests that connectivity disruption in neural circuits is a precursor to neuronal death in Parkinson's disease (PD). 1,6,7 The impairments of mitochondrial dynamics, axonal trafficking, synaptic protein expression, and synaptic integrity are observed before the neurodegeneration in PD. [8][9][10] Furthermore, disrupted connectivities of neural networks, such as synaptic dysfunction or synaptic loss, are the neuropathological hallmarks of early stage PD. 11,12 A strategy to reconnect neural circuits might therefore be able to slow the clinical progression of PD.
During development, multiple guidance cues can control the guidance of axons to their specific targets and help establish functional neural circuits, such as netrin-1 (NTN-1)/deleted in colorectal carcinoma (DCC), Slit3/robo, Semaphorin5A/plexin, and ephtinB/ephB pathways. 13 However, the function of these guidance molecules in the adult brain, particularly during the neurodegenerative process, is unclear. Among these, NTN-1 and its receptor DCC play a critical role in the development and function of the midbrain dopamine circuitry. 14,15 Some studies have shown that these guidance molecules are significant predictors of PD outcomes, 16,17 whereas others found that these guidance molecules have a weak association with PD. 18,19 Recently, some findings provide evidence that the NTN-1/DCC pathway modulates the survival and death of dopamine neurons and may contribute to non-motor and motor symptoms in PD. 20,21 These results confirm a key role of the NTN-1/DCC pathway in adult dopamine neuron fate in PD. However, it is still unknown how the NTN-1/DCC pathway is altered during the onset and progression of PD.
In the present study, we first examined changes in the NTN-1/ DCC pathway in multiple animal models of PD and altered NTN-1/ DCC signaling in SH-SY5Y cells after exposure to 1-Methyl-4-phenyl pyridinium iodide (MPP+). Then, when comparing the plasma level of NTN-1 between PD patients and healthy controls (HC), we investigated the correlation between plasma levels of NTN-1 and clinical symptoms of PD.

| Animals
Experiments were approved by the Institutional Animal Care and Use Committee of Nanjing Medical University and followed the 3R rules. C57BL/6 mice (male, aged 10-12 weeks, weighing 23-25 g) were purchased from the Animal Core Facility of Nanjing Medical University (Nanjing, Jiangsu, China). All mice were housed in five per cage groups in a standardized light-dark cycle at 22°C and were fed standard rodent food and water.  Figure 1A, MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahy dropyridine, Cat#HY-15608, MedChemExpress, Shanghai, China; 25 mg/kg body weight, in saline, s.c.) was administered at 24 h intervals for five consecutive days. The same volume of saline was injected as a control. Blood samples and brains were collected from mice on Days 1 (referred to as MPTP + 1d), 3 (referred to as MPTP + 3d), or 7 (referred to as MPTP + 7d) after the last injection. We performed behavioral tests 7 days after the last injection to verify whether the model mice developed Parkinson-like manifestations. For investigating earlier alterations of NTN-1 and DCC, mice were sacrificed 6 h after 1 (referred to as MPTP-1d), 3 (referred to as MPTP-3d), or 5 (referred to as MPTP-5d) doses of MPTP injection ( Figure 2A).
The respective controls were injected with equivalent volumes of sterile saline. Brain samples were collected on day 7 after LPS injections for the subsequent immunohistochemistry and western blot experiments.

| Chronic MPTP/p model of PD
As shown in Figure S2D, MPTP (25 mg/kg body weight, in saline, s.c.) and a clearance inhibitor probenecid (Cat#HY-B0545, MedChemExpress, Shanghai, China; 250 mg/kg body weight, in saline, i.p.) was administered twice a week for 5 weeks. Brain samples were collected from mice on day 7 after the last injection.

| Open field test
Seven days after the last injection, mice were individually placed in an open field apparatus (50 × 50 cm) for 5 min. The total distance was measured and analyzed using the TopScan system (CleverSys Inc., Reston, USA).

| Accelerating rotarod test
Mice were trained for 3 days on the rotarod at a constant speed of 4 rpm, as previously described. 22 Seven days after the last injection, mice were placed on the rotarod at speeds accelerating from 4 to 40 rpm in 5 min. In three successive trials, the latency to fall from the rotarod was measured, and mean latency values were calculated.

| Pole test
As previously described, 23 the time needed for the mouse to turn completely head downward (T-turn) and the total time until the mouse reached the floor with its four paws (T-total) were recorded. T-turn and T-total were the best performances for each session of the trial.

| Cell culture and treatment
Human SH-SY5Y neuroblastoma cells were cultured as previously described. 24

| Western blot analysis
Brain samples or cells were lysed as previously described. 25 The proteins were denatured in aSn DS sample buffer, separated by 10% SDS-PAGE, and transferred to PVDF membranes. Blots were blocked with 5% skimmed milk in Tris-buffered saline with 0.1% Tween 20 (TBST)

| Statistical analysis
The continuous data were expressed as the mean ± SEM or as the median (first quartile, third quartile), depending on the distribution. Statistical analyses were performed by GraphPad Prism software 9.0 (GraphPad Software, San Diego, CA, USA). Normality was evaluated by the Shapiro-Wilk test. Normally distributed data were analyzed by two-sided t test or one-way analysis of variance (ANOVA) followed by Tukey posthoc test, and nonparametric data were analyzed by Mann-Whitney test. The enumeration data were presented as case numbers or percentages, and two groups were compared using a Chi-squared test. The correlation between plasma NTN-1 levels and clinical factors such as the H-Y stage and UPDRS was analyzed with Pearson correlation. p < 0.05 was considered to be statistically significant.

| An imbalance of NTN-1 and DCC is associated with dopamine neurodegeneration in mice
After five MPTP injections in the subacute model, mice were analyzed on days 1, 3, and 7 after the last MPTP injection ( Figure 1A).
To mimic the slow development of human pathology, we then repeated injections of MPTP 10 times for 5 weeks to create the chronic model ( Figure S2D). The MPTP/p chronic model showed a reduction in TH-positive cells in the SNpc ( Figure S2E) and a decrease in TH protein levels in the midbrain (p = 0.011, two-sided t test, Figure S2F). There was also a dramatic decline in NTN-1 levels, a significant increase in DCC levels, and a decreased ratio of NTN-1 to DCC levels in the SNpc (p = 0.023, two-sided t test, Figure S1F).
These mouse models suggest that the imbalance of the NTN-1/DCC pathway is a common feature of dopaminergic neurodegeneration.

| Alteration of NTN-1 and DCC precedes the death of dopaminergic neurons in the subacute MPTP model
Although the expression of NTN-1 and DCC in substantia nigra has been observed during the late phase of the murine MPTP subacute model, earlier alteration of nigral NTN-1 and DCC in MPTP-treated mice is uncertain. We performed an in situ doubleimmunofluorescent assay for NTN-1 or DCC with a marker for labeling dopaminergic neurons TH over a time course of the MPTP subacute model (Figure 2A). In the saline-treated group, NTN-1 is found predominantly in the cytoplasm of dopaminergic neurons in the substantia nigra ( Figure 2B). NTN-1 expression was significantly reduced as early as 6 h following the first MPTP injection, and this reduction persisted during successive MPTP insults ( Figure 2B).
Then, we sought to identify DCC protein localization in the nigrostriatal pathway. The expression of DCC protein was weak in TH + neurons of the SNpc in the sham group and gradually increased over time during MPTP injections. Especially, DCC protein was expressed highly in the remaining TH + neurons on day 7 after the last MPTP injection ( Figure 2C). The results of the double-immunofluorescence assay indicate that the alterations of NTN-1 and DCC occurred first, before the loss of dopaminergic neurons.

| Dose-and time-dependent alterations of NTN-1 and DCC after MPP+ stimulation in SH-SY5Y cells
To further investigate the relationship between the NTN-1/DCC pathway and cell death, human SH-SY5Y neuroblastoma cells were In the MPP+-treated cells, NTN-1 protein levels were decreased while DCC levels were elevated, resulting in a dose-dependently decreased ratio of NTN-1/DCC (F(5, 18) = 16.37, p < 0.001, one-way ANOVA, Figure 3B). Immunofluorescence results further confirmed the increased expression of DCC in MPP+-treated cells ( Figure 3C).

| Idiopathic PD patients have lower plasma levels of NTN-1
We then conducted correlation studies between plasma NTN-1 and parkinsonian symptoms to understand how this pathway contributes to PD. The demographic data are shown in Table 1. Sex, BMI, and age in study did not significantly differ between idiopathic PD and aged-matched HC. The median concentration of NTN-1 in patients with idiopathic PD (133.6 (77.62-233.1) pg/mL) was significantly lower (p = 0.039, Mann-Whitney test, Figure 6A Figure 6C), II (r = 0.271, p = 0.023, Figure 6D), and III (r = 0.300, p = 0.038, Figure 6E) in idiopathic PD. NTN-1 levels did not correlate with age ( Figure S4A), BMI ( Figure S4B), course of disease ( Figure S4C), and H-Y stage ( Figure S4D). It is worth noting that if the data of only one patient in Stage 4 were removed, the p value of correlation between NTN-1 and H-Y stage would reach 0.058 ( Figure S4E).

| DISCUSS ION
Netrin-1 and its receptor DCC are highly expressed in adult nigral dopamine neurons, 20,26 suggesting that they may be involved in the function or degeneration of these cells. Recently, Jasmin et al.
found that silencing NTN-1 in the adult substantia nigra of mice leads to the cleavage of DCC and loss of dopamine neurons. 20 In the 6-hyroxydopamine-lesioned rat model and A53T-SNCA mouse PD model, a similar therapeutic potential exists for overexpressing NTN-1 and injecting recombinant NTN-1 into the brain. 20 In addition, they found a reduction in NTN-1 expression among postmortem brain samples and gut biopsies from PD patients. 20,27 In the present study, we found that the imbalance of the NTN-1/DCC F I G U R E 3 Dose-dependent alterations of NTN-1 and DCC after MPP+ stimulation in SH-SY5Y cells. (A) Cell viability at 24 h after MPP+ stimulation was measured using the MTT assay. Data were presented as mean ± SEM for four independent experiments with three multiple holes in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. ***p < 0.001, **p < 0.01 versus Con group. (B) Western blotting analysis of NTN-1 and DCC at 24 h after different doses of MPP+ in SH-SY5Y cells. Data were presented as mean ± SEM for four independent experiments in each group and analyzed by one-way ANOVA followed by Tukey posthoc test. ***p < 0.001, *p < 0.05 versus Con group. (C) Representative images of DCC (Green) in SH-SY5Y cells at 24 h after 100 μM or 500 μM MPP+ stimulation. Scale bar, 100 μm. MPP+, 1-Methyl-4-phenylpyridinium iodide; DCC, deleted in colorectal carcinoma; Con, control.
pathway is a common feature of dopaminergic neurodegeneration and, for the first time, revealed that plasma NTN-1 levels were significantly lower in PD patients than in aged-matched HC.
In the present study, we found that the neurodegeneration of nigral DA neurons was always accompanied by an imbalance of NTN-1 and DCC, as shown in MPTP and LPS PD models. Using NTN-1 conditional knockout (KO) mice, Jasmin et al. found that deleting NTN-1 in the SNpc caused severe motor impairment and a loss of nigral DA neurons, accompanied by increased levels of DCC receptors. 20 Further experiments in aged DAT cre /DCC fl/fl mice revealed a significant decrease in ventral midbrain DA neurons. 21 These findings are in line with our speculation that the imbalance between NTN-1 and DCC should be a common feature of DA neuron injury.
Furthermore, we found that subacute MPTP exposure resulted in earlier onset of these imbalances than a reduction of TH expression in the SNpc. In the brain, MPTP is converted to its final toxic metabolite MPP + , which selectively enters DA neurons via the DA transporter and then inhibits Complex I of the elec-

F I G U R E 4 Time-dependent alterations of NTN-1 and DCC after MPP+ stimulation in SH-SY5Y cells. (A)
Cell viability at different times after 100 μM MPP+ stimulation was measured using the MTT assay. Data were presented as mean ± SEM for four independent experiments with three multiple holes in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. ***p < 0.001, **p < 0.01 versus Con group. (B) Western blotting analysis of NTN-1 and DCC at different times after 100 μM MPP+ stimulation in SH-SY5Y cells. Data were presented as mean ± SEM for four independent experiments in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. ***p < 0.001 versus Con group. (C) Representative images of DCC (Green) in SH-SY5Y cells at 48 and 72 h after 100 μM MPP+ stimulation. Scale bar, 100 μm. Con, control; DCC, deleted in colorectal carcinoma; MPP+, 1-Methyl-4-phenylpyridinium iodide.

F I G U R E 5 Dose-and time-dependent alterations of NTN-1-DCC signaling after MPP+ stimulation in SH-SY5Y cells. (A) Western blotting analysis of p-PAK and p-Src at 24 h after different doses of MPP+ stimulation in SH-SY5Y cells.
Data were presented as mean ± SEM for four independent experiments in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. ***p < 0.001, **p < 0.01 versus Con group. (B) Western blotting analysis of Bax, Bcl-2, cleaved caspase-3, and caspase-3 at 24 h after different doses of MPP+ stimulation in SH-SY5Y cells. Data were presented as mean ± SEM for four independent experiments in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. ***p < 0.001, **p < 0.01, *p < 0.05 versus Con group. (C) Western blotting analysis of p-PAK and p-Src at different times after 100 μM MPP+ stimulation in SH-SY5Y cells. Data were presented as mean ± SEM for four independent experiments in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. **p < 0.001, **p < 0.01, *p < 0.05 versus Con group. (D) Western blotting analysis of Bax, Bcl-2, cleaved caspase-3, and caspase-3 at different times after 100 μM MPP+ stimulation in SH-SY5Y cells. Data were presented as mean ± SEM for four independent experiments in each group and analyzed by one-way ANOVA followed by Tukey post hoc test. ***p < 0.001, **p < 0.01, *p < 0.05 versus Con group. Con, control; DCC, deleted in colorectal carcinoma; MPP+: 1-methyl-4-phenylpyridinium iodide; NTN-1, netrin-1.
Focal adhesion kinase and Src play a critical role in netrin signaling in axon outgrowth, migration, and axon attraction. 29 Both FAK and Src stimulate DCC phosphorylation and are potential downstream effects of DCC. FAK has been implicated in the function of netrin through its links to other molecules, such as PLCγ, PI3K, Rac, Cdc42, and MAP kinases. MPP+ can reduce NTN-1 expression, followed by reduced phosphorylation levels of FAK and Src in a concentration-and time-dependent manner, which might explain why axon growth was inhibited before MPP+ depletes cellular ATP in previous studies. 30 Generally, without NTN-1, DCC leads to cell death independent of the mitochondrial pathway and the death receptor TA B L E 1 Clinical characteristics of the patients with PD and healthy controls. In line with previously reported postmortem brain samples from PD patients, we found a reduction in plasm netrin-1 levels in patients compared with aged-matched HC. A reduction in serum NTN-1 levels was also observed in patients with subclinical atherosclerosis, 34 type 2 diabetes mellitus, 35 and multiple sclerosis. 36 Thus, NTN-1, particularly in plasm, seems unable to identify DA neurodegeneration in a specific manner, depending on the specific type of nerve damage and stage of disease development.
In the present study, we discovered correlations between plasma NTN-1 levels and UPDRS but not the H-Y stage or disease course.
The UPDRS is the most widely used scale to measure impairment and disability in PD. 37 Although it has not been designed for classification on levels of severity, the different scores of the UPDRS may suggest categorical levels of disease severity at bedside use. 38 The H-Y stage is another widely used and accepted staging system for the severity of PD, which significantly correlates with both qualityof-life measures and studies of objective motor performance.
Because the H-Y stage is driven mainly by motor features and disability, the proportion of severity levels in UPDRS Parts II and III were highly correlated with the H-Y stage, as previously reported. 39 Although we only found the correlations between plasma NTN-1 shows that serum NTN-1 was associated with recovery after ischemic stroke. 41 NTN-1 can promote neuronal regeneration, reduces ischemia-reperfusion injury, and regulates blood-brain barrier integrity in ischemic lesions. It has been speculated that the increase in plasma NTN-1 may act as a protective mechanism against neurondegeneration in PD. Thus, further investigating should be carried out to clarify the correlation between NTN-1 and PD symptoms, including recruiting a larger sample size of all H-Y stages, analyzing the relationship between circulating NTN-1 levels and non-movement symptoms, and obtaining centrally derived NTN-1 from PD patients. It is needed to determine whether reduced plasma NTN-1 results from PD susceptibility or is a consequence of neurodegeneration.
In conclusion, a significant decrease in plasma NTN-1 was found for the first time in PD patients, with a close correlation to symptoms. Furthermore, as a common feature of nigral DA neuron injury in the early stages, the imbalance of NTN-1 and DCC appears before cell death. However, more studies are needed to clarify the potential differences between central and peripheral NTN-1. In addition, the NTN-1/DCC signaling pathway needs to be tested in models other than MPTP neurotoxin to confirm whether it contributes to neurodegeneration in PD and whether targeting this signaling pathway can alleviate or even reverse the damage of DA neurons. writing, reviewing, and revising the paper. All authors read and approved the final paper.

ACK N OWLED G M ENTS
This study was supported by grants from the National Natural Health Committee (to Y.L.).

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.