Role of upregulation of the KATP channel subunit SUR1 in dopaminergic neuron degeneration in Parkinson’s disease

Abstract Accumulating evidence suggests that ATP‐sensitive potassium (KATP) channels play an important role in the selective degeneration of dopaminergic neurons in the substantia nigra (SN). Furthermore, the expression of the KATP channel subunit sulfonylurea receptor 1 (SUR1) is upregulated in the remaining nigral dopaminergic neurons in Parkinson's disease (PD). However, the mechanism underlying this selective upregulation of the SUR1 subunit and its subsequent roles in PD progression are largely unknown. In 3‐, 6‐, and 9‐month‐old A53T α‐synuclein transgenic (α‐SynA53T+/+) mice, only the SUR1 subunit and not SUR2B or Kir6.2 was upregulated, accompanied by neuronal damage. Moreover, the occurrence of burst firing in dopaminergic neurons was increased with the upregulation of the SUR1 subunit, whereas no changes in the firing rate were observed except in 9‐month‐old α‐SynA53T+/+ mice. After interference with SUR1 expression by injection of lentivirus into the SN, the progression of dopaminergic neuron degeneration was delayed. Further studies showed that elevated expression of the transcription factors FOXA1 and FOXA2 could cause the upregulation of the SUR1 subunit in α‐SynA53T+/+ mice. Our findings revealed the regulatory mechanism of the SUR1 subunit and the role of KATP channels in the progression of dopaminergic neuron degeneration, providing a new target for PD drug therapy.


| INTRODUC TI ON
Parkinson's disease (PD), the second most common age-related human neurodegenerative disease, is characterized by classical motor symptoms such as muscular rigidity, resting tremor, bradykinesia, and postural reflex disorders (Armstrong & Okun, 2020).
The main pathological feature of PD is the selective loss of dopaminergic neurons in the substantia nigra (SN) accompanied by abnormal aggregation of α-synuclein (α-Syn), that is, Lewy bodies (LBs) (Vazquez-Velez & Zoghbi, 2021). In addition to the genetic factors, environmental factors, aging, inflammation, oxidative stress, mitochondrial dysfunction, and abnormal iron metabolism, the selective activation of ATP-sensitive potassium (K ATP ) channels in dopaminergic neurons is also mediated the pathogenesis and development of PD (Abdelkader et al., 2020;Jiang et al., 2017;Nguyen et al., 2019;Trist et al., 2019;Zhang et al., 2018).
Three types of K ATP channels are expressed in the dopaminergic neurons: SUR1/Kir6.2, SUR2B/Kir6.2, and SUR1/SUR2B/Kir6.2 (Liss et al., 1999). The co-expression of SUR1 and Kir6.2 in functional K ATP channels is highly sensitive to metabolic inhibition in the SN (Yamada et al., 1997). Interestingly, the mRNA levels of the SUR1 subunit were found to be approximately twofold higher in dopaminergic neurons from patients with PD than in those from individuals in the control group, as determined by quantitative mRNA expression profiling techniques (Schiemann et al., 2012). In addition, previous investigations suggested that the activity of K ATP channels promotes the switch from tonic firing to N-methyl-D-aspartate receptor (NMDA-R)-mediated burst activity in vivo (Schiemann et al., 2012;Zweifel et al., 2009). Enhancement of K ATP channel gating was necessary for burst firing in dopaminergic neurons in the SN. In addition, the dopaminergic neurons in the SN in awake PD patients exhibited high levels of burst firing activity, which accelerated the degeneration of dopaminergic neurons (Dragicevic et al., 2015; W. Lin et al., 2006;Liss et al., 1999;Liss et al., 2005). Given the abnormal expression and the role of K ATP channels in patients with PD, it is highly important to clarify whether the expression of K ATP channels in nigral dopaminergic neurons is changed in the early stage of PD and whether the K ATP channel dysfunction is a cause or a consequence of the dopaminergic neuron degeneration.
The SUR1 subunit can be regulated by heat shock proteins, insulin, ghrelin, and protein kinase A in the peripheral nervous system (Chen et al., 2013;Heddad Masson et al., 2014;Yan et al., 2010).
The SUR1 subunit was found to be upregulated at 1 hr and 8 hr after spinal cord injury, and this upregulation was alleviated by activation of growth hormone secretagogue receptor-1a (Lee et al., 2014). In addition, FOXA1 and FOXA2, members of the forkhead/winged helix transcription factor family, were demonstrated to be able to bind to the SUR1 promoter directly in pancreatic beta cells (Cirillo et al., 2002;Jackson et al., 2010). However, little is known about the regulatory mechanism of SUR1 expression in nigral dopaminergic neurons.
In the present study, using mutant human A53T α-Syn transgenic (α-SynA53T +/+ ) mice, we aimed to investigate the changes in the expression and the role of K ATP channels in nigral dopaminergic neuron degeneration in PD. To further explore the underlying mechanisms of SUR1 subunit dysregulation, we assessed the expression of FOXA1 and FOXA2 and their interactions with the SUR1 subunit.
Our findings provide new insight into the involvement of K ATP channels in the progression of PD.
Most importantly, both the mRNA and protein levels of the K ATP channel SUR1 subunit were increased, which might influence the function of dopaminergic neurons in PD.
We refer to the criterion of dopaminergic neurons' firing pattern defined by Grace and Bunney, which is explained in detail in our Methods section (Grace & Bunney, 1984a, 1984b. In our analysis of firing patterns (Figure 2j), in 3-month-old α-SynA53T +/+ mice, 6 of the 11 (54.55%) dopaminergic neurons were identified as regular-firing neurons, 4 of the 11 (36.36%) dopaminergic neurons were identified as irregularly discharging neurons, and 9% of the dopaminergic neurons were identified as cluster discharging neurons. In 6-month-old α-SynA53T +/+ mice, 6 of the 11 (54.55%) dopaminergic neurons were identified as irregularly discharging neurons, and 5 of the 11 (45.45%) dopaminergic neurons were identified as cluster discharging neurons. In 9-month-old α-SynA53T +/+ mice, 9 of the 13 (69.23%) dopaminergic neurons were identified as cluster discharging neurons. Consistent with the analysis of the proportion of spikes in bursts, the percentage of cluster discharging neurons was increased in 6-month-old α-SynA53T +/+ mice (chisquare, p = 0.013) and 9-month-old α-SynA53T +/+ mice (chi-square, F I G U R E 2 Spontaneous firing activities of dopaminergic neurons in the SN of α-SynA53T +/+ mice at different ages. (a, d, g) The original spikes of both WT and α-SynA53T +/+ mice in 3-, 6-, and 9-month, respectively. (b, e, h) The summarized data showing the spontaneous firing rate of dopaminergic neurons in SN of both WT and α-SynA53T +/+ mice in 3-, 6-, and 9-month, respectively. (c, f, i) The summarized data showing the proportion of spikes in bursts of both WT and α-SynA53T +/+ mice in 3-, 6-, and 9-month, respectively. (j) The summarized data showing the percentage of the three firing patterns in WT and α-SynA53T +/+ mice in 3-, 6-, and 9-month ( * p < 0.05, *** p < 0.001) p = 0.010) compared with the age-matched WT mice. These results indicated that the firing rate and firing pattern of dopaminergic neurons changed with age, as especially evident in 9-month-old α- Moreover, in order to verify if the changes of the firing activity were affected by the SUR1, we also observed the spontaneous firing frequency and the proportion of spikes in bursts after α-Syn knockdown in mice. As shown in Figure S2a-c, the spontaneous firing frequency of A53Tα-Syn-KD mice was not changed (A53T: 3.84 ± 0.62 Hz, A53Tα-Syn-KD: 3.88 ± 0.36 Hz, n = 11,12, unpaired t test, p = 0.948) but the proportion of spikes in bursts was decreased significantly (A53T: 47.55 ± 8.56%, A53Tα-Syn-KD: 6.99 ± 1.39%; n = 11,12; unpaired t test, p < 0.0001). In summary, the firing activity of dopaminergic neurons was affected in the progression of PD with α-Syn overexpression.

| The K ATP channel inhibitor glibenclamide increased the frequency of spontaneous discharge of nigral dopaminergic neurons in mice at different ages
To explore whether K ATP channels intimately affect the activity of dopaminergic neurons in PD, we used glibenclamide as a K ATP channel inhibitor in vivo. Firstly, in 3-month-old α-SynA53T +/+ mice, the firing rate of nigral dopaminergic neurons was increased 106% after glibenclamide treatment (basal: 2.26 ± 0.80 Hz, glibenclamide: 4.66 ± 1.70 Hz; n = 8; paired t test, p = 0.0023), as shown in

| FOXA1 and FOXA2 could act as transcription factors to regulate the expression of the SUR1 subunit
To verify whether the upregulation of SUR1 expression was related to FOXA1 and FOXA2, we first examined the alterations in the mRNA levels of FOXA1 and FOXA2 in the SN in α-SynA53T +/+ mice. Figure 5a,b, the mRNA levels of FOXA1 and FOXA2 in the SN in 3-month-old α-SynA53T +/+ mice were increased by 37.27% (unpaired t test, p = 0.0101) and 153% (unpaired t test, p = 0.0206), respectively, compared with those in the age-matched WT mice.

As shown in
In 6-month-old α-SynA53T +/+ mice, the mRNA levels of FOXA1 and FOXA2 in the SN were increased by 180% (unpaired t test, p = 0.0233) and 204% (unpaired t test, p = 0.0083), respectively, compared with those in the age-matched WT mice. In 9-month-old α-SynA53T +/+ mice, the mRNA levels of FOXA1 and FOXA2 in the SN were increased by 130% (unpaired t test, p = 0.0130) and 179% (unpaired t test, p = 0.0060), respectively, compared with those in the age-matched WT mice.
We also detected the expression of the transcription factors F I G U R E 3 Effects of glibenclamide on spontaneous firing activities of dopaminergic neurons in the SN of α-SynA53T +/+ mice at different ages. (a, e, and i) Typical frequency histogram showed that the firing rate of dopaminergic neurons was increased by glibenclamide in the SN in 3-(a), 6-(e), and 9-month α-SynA53T +/+ mice (i). (c, g, and k) The original spikes at different stages in 3-(c), 6-(g), and 9-month α-SynA53T +/+ mice (k). (b, f, and j). The summarized data showing the effects of glibenclamide on the firing rate of dopaminergic neurons in the SN in 3-, 6-, or 9-month α-SynA53T +/+ mice. (d, h, and l). The summarized data showing the effects of glibenclamide on the proportion of spikes in bursts of dopaminergic neurons in the SN in 6-and 9-month α-SynA53T +/+ mice ( ** p < 0.01, respectively, compared with that in MES23.5 cells transfected with only A53Tα-Syn (p < 0.05, p < 0.001).
Then, we performed a dual-luciferase reporter assay to verify whether SUR1 was regulated by FOXA1 and FOXA2 (Figure 5h,i).

| DISCUSS ION
In the present study, we demonstrated that FOXA1 and FOXA2 were the causes of upregulation of SUR1 subunit in α-Syn overexpressed models in vivo and in vitro. Moreover, the changes in electrical activities regulated by K ATP channels play an important role in the progression of dopaminergic neurons degeneration in PD.
We previously demonstrated that the dopamine transporter (DAT) levels in α-SynA53T +/+ mice were decreased at the ages of 6, 9, and 12 months by 11 C-2β-carbomethoxy-3β-(4-fluorophenyl) tropane positron emission tomography ( 11 C-CFT PET), indicating that the ability of DAT in dopaminergic neurons was reduced. However, F I G U R E 4 Effects of SUR1 knockdown on the progression of dopaminergic neurons degeneration in the SN of α-SynA53T +/+ mice. (a). Schematic diagram of virus injection location. (b, c). The SUR1 protein levels of α-SynA53T +/+ mice with LV-SUR1-RNAi in the SN were decreased, compared with the control group. (d, e). The change of TH protein levels after SUR1 knockdown in α-SynA53T +/+ mice. (f, g). The change of the number of THpositive neurons after SUR1 knockdown in α-SynA53T +/+ mice ( * p < 0.05, ** p < 0.01). Scale bar = 100 μm the DAT levels were not significantly decreased in 3-month-old α-SynA53T +/+ mice . The decreased DAT binding ability was associated with the progression of PD patients (Ikeda et al., 2019;Ishibashi et al., 2014). According to the previous studies, we examined the expressions of K ATP channels subunits in α-SynA53T +/+ mice at the ages of 3, 6, and 9 months. It is noteworthy that the expression of the SUR1 subunit has been increased in 3-month-old α-SynA53T +/+ mice, even though the mice did not show significant loss of dopaminergic neurons at that age. A higher level of SUR1 expression has been reported to promote the K ATP channels assembling and cell membrane trafficking, enhance the open abilities of the channels on cell surface (Yan et al., 2010).
General consensus is that K ATP channels could control the neuronal excitability, energy metabolism, and neurotransmitter release (Mollajew et al., 2013;Nelson et al., 2015;Nguyen et al., 2019). But in different ages of α-SynA53T +/+ mice, we only observed changes in the autonomous firing rate in 9-month-old α-SynA53T +/+ mice, and we did not observe a significant difference in the autonomous firing rate in 3-or 6-month-old α-SynA53T +/+ mice. Moreover, the addition of glibenclamide could increase the firing rate in 3-and 6-month-old α-SynA53T +/+ mice. Consistent with our study, Lasser- (e, f). The mRNA expression of FOXA1 (e) and FOXA2 (F) after MES23.5 cells transfected with WTα-Syn and A53Tα-Syn for 24 h. The house-Keeping gene, GAPDH, was served as the standardized control. (g) Change of the mRNA levels of SUR1 in A53T α-Syn group after the MES23.5 cell transfected with siFOXA1 + A53Tα-Syn or siFOXA2 + A53Tα-Syn for 24 h. (h, i) Dual luciferase results showed that the firefly/ renilla luciferase activity in the SUR1-FOXA1 group (h) and SUR1-FOXA2 group (i) ( * p < 0.05, ** p < 0.01, *** p < 0.001) whole-cell conductance and decreased the firing rate in single dopaminergic neurons in the SN, and these effects were diminished by inhibition of K ATP channels. This discrepancy might arise from the difference of the pathogenesis mechanism between α-Syn oligomers treatment in vitro and mutant α-Syn overexpression in α-SynA53T +/+ mice; alternatively, the regulation of circuits in vivo could also be a cause. Therefore, we hypothesized that the different results in α-SynA53T +/+ mice at different ages might be caused by the balance of mutant α-Syn overexpression and the function of K ATP channels.
However, the opening of K ATP channels could induce hyperpolarization of the cell membrane and decrease the autonomous firing rate by inhibiting Ca 2+ flux. A similar phenomenon has been observed in hippocampal neurons, suggesting that K ATP channels could dampen the enhanced neuronal excitability induced by the formation of nonselective cation channels by α-Syn (Mironov, 2015). Therefore, in the early stage of PD, like in 3-and 6-month-old α-SynA53T +/+ mice, we did not observe any changes in the spontaneous firing rate of dopaminergic neurons, the K ATP channels attempted to rectify the hyperactivity of dopaminergic neurons induced by α-Syn overexpression (Hill et al., 2021). However, in 9-month-old α-SynA53T +/+ mice, the firing rate of the dopaminergic neurons began to increase with the gradual aggregation of α-Syn.
But the effects of K ATP channel activity might be a double-edged sword in PD. Previous studies have indicated that the discharge pattern of remaining dopaminergic neurons in PD patients is changed from a tonic single peak discharge pattern to a cluster discharge pattern, and in which NMDA receptors and K ATP channels are essential components (Dragicevic et al., 2015;Schiemann et al., 2012;Shen & Johnson, 2010;Shen et al., 2014). In our studies, we also verified that K ATP channels participate in NMDA-mediated burst firing in dopaminergic neurons in vivo in 3-month-old WT mice (as shown in Figure S3). α-Syn could enhance the activation effect of Tau on nonreceptor tyrosine kinase Fyn by combining with Tau, inducing Fyn overexpression. Moreover, Fyn could phosphorylate the c-terminal site Y1325 of NR2A receptor, which leads to the hyperactivation of NMDA receptors (Trudler et al., 2021). Therefore, the occurrence of burst firing should be partially blocked by the inhibition of K ATP channels, but in α-SynA53T +/+ mice, with α-Syn overexpression induced high spontaneous firing rate, the occurrence of burst firing was not changed. However, α-Syn knockdown may affect the expression or the function of K ATP channels and NMDA receptors at the same time. Therefore, we observed the occurrence of burst firing was decreased significantly after α-Syn knockdown in the present study (as shown in Figure S2a,b). The hyperexcitability and burst firing of neurons could promote DA release in the short term, but long-term high-frequency firing would lead to intracellular Ca 2+ overload (Beatty et al., 2012;Shen & Johnson, 2010). Even worse, this vicious circle could eventually induce the degeneration of dopaminergic neurons. To further verify the role of K ATP channels in the degeneration of dopaminergic neurons in PD, lentivirus-mediated interference with SUR1 subunit expression in the SN was accomplished in 4-month-old α-SynA53T +/+ mice. After 2 months, both the TH protein expression level and the number of TH-positive neurons were significantly increased, indicating that interference with SUR1 expression at a very early stage of PD can partially antagonize dopaminergic neuron degeneration and delay the progression of PD.
Although some regulatory factors of SUR1 subunit have been demonstrated, the underlying regulatory mechanism of SUR1 subunit in nigral dopaminergic neurons is still largely unknown. FOXA1 and FOXA2 have been proven to play an essential role in the generation of dopaminergic neurons during early and late stages of development (Ang, 2009;Lin et al., 2009). Although the synthesis and release of DA in the SN was decreased in FOXA1/2 −/− mice at 12 weeks of age, the number of TH-positive neurons in FOXA1/2 −/− mice at 24 weeks of age did not change (Domanskyi et al., 2014;Pristera et al., 2015;Stott et al., 2013). FOXA1 and FOXA2 can also regulate the expression of the SUR1 subunit as transcription factors in β cells (Cirillo et al., 2002;Jackson et al., 2010). In the present study, the levels of FOXA1 and FOXA2 were significantly increased in 3-, 6-, and 9-month-old α-SynA53T +/+ mice and in α-Syn-overexpressing cells. Furthermore, we observed that FOXA1 and FOXA2 directly bound to the promoter region of SUR1. FOXA1 and FOXA2 might be mediators regulating the upregulation of SUR1 expression in α-Syn overexpression models. However, the relationships between α-Syn and FOXA1/FOXA2 need further clarification.
In conclusion, the SUR1 subunit of K ATP channels was upregulated at the early stage of PD, an effect that might be related to the transcription factors FOXA1 and FOXA2. The enhanced activity of K ATP channels antagonized the increase in the autonomous firing rate induced by α-Syn overexpression at the early age, but the effect of K ATP channel activity was weakened with further aggregation of α-Syn. In addition, burst firing was increased with upregulation of the SUR1 subunit, which finally promoted dopaminergic neuron degeneration. The present study might provide evidence for the role of the SUR1 subunit of K ATP channels in the neurodegenerative progression of PD.

| Animals
The transgenic mice expressing A53T human α-syn (α-SynA53T +/+ mice) were originally obtained from Jackson Laboratory (JAX004479, USA). All the mice were housed in their cages with a 12 h light/dark cycle with free access to food and water. Male homozygous (3-, 6-, and 9-month) α-SynA53T +/+ transgenic mice and their wild type littermates were used for the following experiments. histogram and auto-correlogram analysis. The regular firing pattern was characterized by a normal distribution of ISI and at least three identifiable peaks in an auto-correlogram. The irregular firing pattern showed a Poisson distribution of ISI and less than two peaks in an auto-correlogram. The burst firing pattern displayed a positive skewed distribution of ISI and an auto-correlogram with a single initial peak or without peak. The proportion of spikes in bursts was used to measure the burst firing. The onset of the burst was defined as the occurrence of two spikes with an ISI of <80 ms, and the termination of the burst defined as the-occurrence of an ISI of 160 ms (Grace & Bunney, 1984b). After the electrophysiological study, the brains of mice were frozen and cut to examine the sites of recordings.
SN were microdissected in ice-cold ACSF from the 200 μm tissue slices.

| Immunofluorescent staining
Six-month-old mice with 50 mg/kg sodium pentobarbital anesthe- units/mL of streptomycin in a humidified atmosphere containing 5% CO 2 at 37°C. For experiments, cells were seeded at a density of 1 × 10 5 /cm 2 in plates and grown to 70%-80% confluency. MES23.5 cells with A53T mutant α-syn over expressing were cultured in sixwell plates at a density of 1 × 10 5 /cm 2 in plates and grown to 70%-80% confluency for 24 h. The cells were separated to three groups: GV230 vector group, WT α-Syn group, and A53T α-syn group. In the control group, GV230 vectors were transfected into MES23.5 cells alone. In the over-expression group, cells were transfected with the WTα-syn or A53Tα-syn in a serum-free medium. This relatively high efficiency of infection of MES23.5 cells was harvested for the further studies. All vectors were transfected using Lipofectamine 2000. For optimal transfection efficiency, five volumes of DNA relative to lipofectamine 2000 were used.
HEK293 cells' culture method is similar to MES23.5 cells.
HEK293 cells were transfection with Lipofectamine ™ 2000, and the double luciferase reporter gene assay (Promega, E1910) was performed 24 h later. The relative fluorescence intensity was calculated using firefly fluorescence value/renin fluorescence values.

CO N FLI C T O F I NTE R E S T
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available upon request.