The effect of electroacupuncture at ST25 on Parkinson's disease constipation through regulation of autophagy in the enteric nervous system

The effectiveness and safety of electroacupuncture (EA) for constipation have been confirmed by numerous clinical studies and experiments, and there are also studies on the efficacy of EA for Parkinson's disease (PD) motor symptoms. However, there are few researches on EA for PD constipation. Autophagy is thought to be involved in the mechanistic process of EA in the central nervous system (CNS) intervention in Parkinson's pathology. However, whether it has the same effect on the enteric nervous system (ENS) has not been elucidated. Therefore, we investigated whether EA at Tianshu (ST25) acupoint promotes the clearance of α‐Syn and damaged mitochondria aggregated in the ENS in a model of rotenone‐induced PD constipation. This study evaluated constipation symptoms by stool characteristics, excretion volume, and water content, and the expression levels of colonic ATG5, LC3II, and Parkin were detected by Western Blot (WB) and Real‐Time Quantitative PCR (RT‐qPCR). The relationship between the location of α‐Syn and Parkin in the colonic ENS was observed by immunofluorescence (IF). The results showed that EA intervention significantly relieved the symptoms of rotenone‐induced constipation in PD rats, reversed the rotenone‐induced down‐regulation of colonic ATG5, LC3II, and Parkin expression, and the positional relationship between colonic α‐Syn and Parkin proved to be highly correlated. It is suggested that EA might be helpful in treating PD constipation by modulating Parkin‐induced mitochondrial autophagy.

Parkinson's disease (PD) is the second most common and fastest-growing neurodegenerative disorder that affects over 6.2 million people worldwide and this number is expected to increase to more than 12 million by 2040 (Feigin et al., 2021;GBD Neurology Collaborators, 2019; GBD Parkinson's Disease Collaborators, 2018).PD shows motor and non-motor symptoms with the latter seen in more than 90% of PD patients, and 80-90% of them develop constipation, one of the most important nonmotor symptoms that occurs in the early stage of the disease and persists throughout the development of PD, greatly affecting patients' quality of life (Fasano et al., 2015).A population-based study indicated that the risk of PD increased with the severity of constipation (hazard ratio ranged from 3.3 to 4.2) (Lin et al., 2014).Constipation not only puts individuals at increased risk but also represents one of the earliest problems of autonomic dysfunction in PD.Gastrointestinal symptoms have been shown to precede motor symptoms by Parkinson himself (Parkinson, 2002).According to several prospective studies (Cersosimo & Benarroch, 2012;Noyce et al., 2012;Postuma et al., 2013), constipation may occur many years before the onset of motor symptoms in PD (as shown in a regression analysis, constipation is significantly abnormal at least 5 years before the onset of the disease, with an estimated antecedent interval of 15.3 years) and can arise at all stages of PD.Constipation symptoms may worsen as the disease progresses (Jost & Schrank, 1998).
Although constipation is the most common nonmotor symptom in PD patients, clinical trials evaluating the efficacy of various medications have been scarce.At present, PD constipation is mainly addressed with symptomatic treatment strategies, and medications commonly used on the clinical front have significant side effects and limitations (Eichhorn & Oertel, 2001;Jost, 2010;Lembo, Johanson, et al., 2011;Lembo, Schneier, et al., 2011;Ondo et al., 2012;Sullivan et al., 2006) (Table 1).5-HT4 receptor agonists such as tegaserod are promising in reducing constipation symptoms in PD patients, but their clinical application is limited by their significant cardiovascular side effects and drug safety (Sullivan et al., 2006).Therefore, improving intestinal motility in PD patients has become an urgent task.Traditional Chinese medicine (TCM) enjoys a unique physiology, pathology, and disease prevention and control system, which may provide new ideas for resolving the dilemma of PD constipation treatment.PD constipation is due to the loss of transit capability in the large intestine whose functions are closely related to the spleen and stomach.
In TCM, PD constipation is characterized by deficiency of liver-yin and kidney-yin and insufficiency of Qi and blood.As the source of acquired constitution, the spleen and stomach can produce Qi and blood and nourish the liver and kidney.Therefore, regulating Qi and blood of the spleen and stomach is critical for treating PD constipation.
According to TCM, acupuncture regulates Qi and blood, dredges meridians and collaterals, and balances Yin and Yang.Based on the theory of meridians and collaterals, acupuncture does not only work for relieving constipation but also excels at combating brain diseases, making it suitable for treating PD constipation.Located on both sides of the navel with the small intestine in the deep of it, Tianshu (ST25) acupoint is mainly used to treat gastrointestinal diseases.
Both ancient Chinese medicine documents and modern studies have shown that acupuncture has a good therapeutic effect on constipation.Acupuncture, as an important means of TCM therapy, is outstanding in regulating the spleen and stomach.ST25, Zusanli (ST36), and Shangjuxu (ST37) acupoints are often used in the study of gastrointestinal motility regulation.The Yangming Stomach Meridian of Foot has the characteristics of being full of Qi and blood, which is often used to treat spleen and stomach diseases.In terms of meridian circulation and indications, the Yangming Stomach Meridian of Foot is closely related to brain and mental disorders.This coincides with various modern medical doctrines based on the brain-gut axis, such as the microbiota-intestine-brain axis, which may provide a new path to treat cerebrointestinal syndromes, such as PD constipation.Acupuncture has been used to manage constipation due to various causes (Zhu et al., 2021) such as, PD (Yannan et al., 2022).A large-sample multicenter randomized controlled trial (RCT) (Liu et al., 2016) showed that 8 weeks of electroacupuncture (EA) treatment could significantly increase the number of complete spontaneous bowel movements per week compared to sham treatment, and the effect lasted for at least 20 weeks, indicating the efficacy of EA in treating severe chronic functional constipation.Although the effect of acupuncture on gastrointestinal motor function has been validated by numerous clinical trials and animal testing, and there are studies on its efficacy for PD motor symptoms, studies on the effects and mechanisms of acupuncture in the treatment of PD constipation are still insufficient.
Lewy pathology (LP) is a common feature of multisystem neuropathology in PD, including pathology in the enteric nervous system (ENS) (Manfredsson et al., 2018).Neuronal LP is mainly characterized by the formation of Lewy bodies (LBs) and Lewy neurites (LNs) in neuronal cells, and the abnormal aggregation of alpha-synuclein (α-Syn) is the main component of LBs and LNs (Lama et al., 2021).α-Syn is encoded by human alpha-synuclein (SNCA) and is mainly distributed in presynaptic nerve terminals.It interacts with phospholipids and other synapse-associated proteins to promote the release of various neurotransmitters, including dopamine.Under normal conditions, α-Syn maintains the physical connection between the endoplasmic reticulum and mitochondria.However, the abnormal aggregation of α-Syn in LBs and its reduced affinity for the lipid raft region after mutation lead to mitochondrial dysfunction and inhibit mitochondrial repair through autophagy, which eventually induces neurodegeneration.Tian et al. (2016) found that EA could activate the autophagic lysosomal pathway in central neurons to promote the clearance of phosphorylated α-Syn (pα-Syn), thereby restoring dopaminergic neuronal activity in the substantia nigra pars compacta (SNpc).The scavenging effect of EA on α-syn was also verified in ENS.Yannan et al. (2022) demonstrated that EA improved constipation symptoms in mice with PD by reducing α-Syn aggregation in colonic tissue.However, the exact pathway through which EA affects the aggregation of α-syn in the ENS remains unclear.In addition, does EA protect ENS through other means?And how EA affects the interrelationship between mitochondrial damage, α-syn aggregation, and protein degradation system dysfunction in the ENS?
Mitochondrial dysfunction caused by accumulation of α-syn plays an important role in PD pathology (Dexter et al., 1989).Mitochondrial autophagy is a regulatory mechanism by which the body maintains a certain level of mitochondrial quality and quantity in the cell (Matheoud et al., 2016).Activating the mitochondrial autophagy pathway in the presence of damaged mitochondria has a protective effect on the cell by promptly removing the damaged mitochondria (Sliter et al., 2018).Pink1/Parkin is an important pathway for intracellular regulation of mitochondrial autophagy: phosphatase and tensin homolog (PTEN)-induced putative kinase 1 (Pink1) is an important mitochondrial autophagy inducible molecule that reverses the mitochondrial dysfunction caused by high pα-Syn expression and reduces cellular damage (Huang et al., 2017;Liu et al., 2017).Further studies have shown that Pink1 recruits Parkin, an essential inducer of mitochondrial autophagy (Ivankovic et al., 2016), which is an important component of several E3 ubiquitin ligase complexes participating as part of the ubiquitin-proteasome system and mediating protein degradation.In patients with early-onset PD, Pink1/ Parkin-related gene mutations accounted for a high proportion, and Pink1 and Parkin expression deficiency in LBs has been found in PD patients and animal models.EA can partially regulate Pink1-Parkin signaling and autophagy-related pathways.Previous studies have shown that EA can prevent and treat PD by modulating Parkin signaling (Yanchun et al., 2013) and improving the function of the ubiquitin-proteasome system (Deng et al., 2015).
Based on the above evidence, we hypothesized that acupuncture might promote the clearance of α-Syn and damaged mitochondria through modulation of colonic autophagy to treat PD constipation.

| Establishment of the experimental animal model
In this study, 8-week-old Sprague Dawley (SD) rats were supplied by the Beijing Vital River Laboratory Animal Technology Co., Ltd.(No. 110011220101889264, under grant SCXK(JING)2021-0011). Lab rats were confined in an environment with stable parameters (conditions: 12-/12-h ± 1-h light/dark cycle; temperature, 22 ± 2 C; relative humidity 60% ± 5%).The animals were randomly numbered and divided into three groups: PD constipation model group, EA group, and control group (for convenience, the following texts refer to them as PD, EA, and SH groups, respectively), with six animals in each group.They were kept in cages of the same size in groups and had free access to food and water.PD was induced by giving a low dose of rotenone.The PD and EA groups were injected subcutaneously with rotenone solvent on the back of the neck at a dose of 0.1 ml/kg once a day for 5 days a week.The solvent was prepared by dissolving 200 mg of rotenone (M6209; Abmole Bioscience Inc., Houston, TX, USA) in 3 ml of dimethyl sulfoxide (DMSO, D8370; Beijing Solarbio Science & Technology, Tongzhou, Beijing, China), and then fixed to 100 ml with sunflower oil to make up 2 mg/ml of rotenone sunflower oil solvent.The SH group was injected with an equal volume of solvent mixture (3% DMSO sunflower oil solvent).
Weekly metabolic cages were performed after rotenone injection to measure wet stool weight, dry and wet weight difference (ΔW), and stool count (the specific methods of metabolic cage method and stool collection and analysis will be elaborated in Section 2.3.2).The modeled rats were evaluated after 4 weeks of rotenone treatment.Rats with defecation indexes (wet stool weight, ΔW, and stool count) lower than the SH group mean and PD behavioral score ≥ 2 were considered to meet the criteria.The remaining unmodeled rats in the PD and EA groups were killed by cervical dislocation.The process of the animal test is shown in Figure 1.All experiments were approved by the Scientific Investigation Board of the Nanjing University of Traditional Chinese Medicine, Nanjing, China (permission no.202112A047) and performed as per the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals published by the National Science Council, China (under grant 202006A016).

| Acupuncture intervention
EA intervention was performed on rats in the EA group using our previous method (Wang et al., 2021;Xu et al., 2021).After gas anesthesia with isoflurane (4-5% for induction and 1-2% for maintenance; 9020000522; Shenzhen Ruiwode Lift Technology), EA treatment on bilateral ST25 (Tianshu, located 5 mm lateral to the intersection between the upper 2/3rd and the lower 1/3rd in the line joining the xiphoid process and the upper border of the pubic symphysis) was administered to the rats in the EA group.Meanwhile, the rats in the PD group received the same anesthesia but without EA treatment.For EA operation, a pair of disposable stainless steel acupuncture needles (0.2 mm in diameter; 20162270970, Suzhou HUATUO Medical Instruments, Suzhou, Jiangsu, China) were inserted at a depth of 5 mm into the ST25 acupoint and the needles were connected to the HANS-100A instrument (HAN ACUTENS WQ1002F; Beijing Anlong Photoelectric Technology, Haidian, Beijing, China) set to a continuous-wave stimulation with a current of 2 mA and an alternating frequency of 2/15 Hz.All operations were performed five times a week by experienced and practicing acupuncturists and lasted for 4 weeks.

| PD behavioral evaluation
PD behavioral evaluation was performed according to the criteria Chen et al. (2008) developed.The specific scoring criteria are shown in Table 2.Only rats with a behavioral score of ≥2 were included in the next step of screening for defecation indicators.Because rats with a behavioral score of ≥2 have relatively significant neurological deficits, the animal models of PD in this range are more reliable (Chang et al., 2011).

| Stool collection and analysis
Rat stools were collected using the metabolic cage method as follows.The rats were placed in individual wire cages with a separation device, which consisted of two layers of separation nets, the upper layer of which separated the rats from the stools to ensure that the stools would not affect the rats' activities, and the lower layer of which separated the stools from the urine to ensure that the wet stool weight would not be affected by the urine.Each week after the first 5-day treatment, rats were placed in the metabolic cages at 7 p.m. on both Friday and Saturday and their stools were collected at 7 a.m. on both Saturday and Sunday, allowing for a 12-h interval to reduce the error caused by evaporation.
The stools were collected in sealed bags in the same order as the rats were placed.The wet stools were promptly weighed using an analytical balance (QUINTIX313-1CN; Sartorius Scientific Instruments, Shunyi, Beijing, China), and the number of them was counted.Then, the stools were dried in a dryer (70 C, 6 h; SepTree ST-06; Xinchi e-commerce, Foshan, Guangdong, China) until the weight no longer changed.Finally, the dry stools were weighed, and the difference between wet and dry weight (ΔW) was calculated.The dry-wet difference of the stools is expressed as: Wet weight (g) À Dry weight (g).

| Western blot analysis
Protein expression was measured using the method we described previously (Xu et al., 2021).For gel electrophoresis, 10 μg protein was extracted from each sample.
F I G U R E 1 Schedule of the experimental procedures.EA, electroacupuncture; ST25, Acupoint Tianshu; ΔW, dry and wet weight difference of the stools Membranes were blocked and probed with primary and secondary antibodies by the manufacturers' suggested concentrations.The primary antibodies were diluted as listed in Table 3.The secondary antibodies used were anti-rabbit IgG, HRP-linked antibody (1:3000, 14,708, CST), and anti-mouse IgG, HRP-linked antibody (1:3000, 14,708, CST).Please refer to our previously published literature for other specific methods, reagents and instruments, and parameter settings.The gray values of the protein bands were quantified with Image J (NIH, Bethesda, MD, USA).

| Real-time quantitative polymerase chain reaction
We placed the tissue (50-100 mg) in 1 ml of Trizol, ground it with a steel ball, and set the centrifuge at 12,000 rpm for 15 min at 4 C.Then we took the supernatant and let it stand for 5 min before adding 200 μl of chloroform, shaking vortex for 30 s and letting it stand for 5 min.We set the centrifuge at 12000 rpm for 15 min at 4 C, put the supernatant in a new EP tube, added 500 μl of isopropanol, mixed it thoroughly, shook it for 2 min and let it stand for 10 min.We then set the centrifuge at 12,000 rpm for 15 min at 4 C and discarded the supernatant.We added 1 ml of 75% ethanol, mixed the precipitate, set the centrifuge at 7500 rpm for 5 min at 4 C. Then we carefully aspirated the supernatant, set the centrifuge at 4 C, 7500 rpm for 30 s, aspirated the supernatant again, put it under the ventilation of an ultraclean table, and let it dry for 10 min before adding an appropriate amount of DEPC water to dissolve the precipitate (incubate at 55 C for 5 min).With the determined RNA concentration, we prepared the following reverse transcription system in an ice-bath nuclease-free centrifuge tube: UEIrisII RT MasterMix (5X) 4 μl, template RNA 50 ng-1 μg, dsDNase 1 μl, RNase-free water fixed to 20 μl.We mixed it gently, put it in the centrifuge instantaneously (so that the liquid could sink to the bottom) before performing the reaction on a PCR instrument under the following conditions: 2 min warming at 37 C, removal of genomic contamination at 55 C, and 10 min incubation at 85 C (if the target RNA is less than 3 kb, the reaction time may be shortened to 5 min), 10 s.We then placed the obtained product quickly on ice or stored it immediately at À20 C for subsequent experiments.0.2 ml PCR tubes were used to prepare the following reaction system: 2Â qPCR Mix 10 μl, 10 μm forward primer 0.4 μl, 10 μm reverse primer 0.4 μl, reverse transcription product 2.0 μl, ddH2O 7.2 μl.three tubes were prepared for each reverse transcription product.The sequences for the primer sets are shown in Table 4. PCR Amplification: predenaturation at 95 C for 5 min; a total of 40 PCR cycles at 95 C first for 10 s and then at 60 C for 30 s; melting curve (60 C-95 C), and 1.6 C temperature rise per second (60 C, 1 min to 95 C, 1 s).The ΔΔCT method was used to calculate the relative expression.

| Immunofluorescence staining
The method we described earlier (Xu et al., 2021) was adopted.The colonic tissue was stripped of the outer connective tissue on the ice immediately after acquisition and unfolded along the longitudinal axis.We flattened and washed the contents in saline before fixing them.
The connective tissue was removed as clean as possible, otherwise, it would affect the results.The dilution of the primary antibody was 1:1000 (Abcam ab212184, CST 4211).The secondary antibodies used were Alexa Fluor 488 (goat anti-rabbit, 1:500, Abcam, Cambridge, UK), and Alexa Fluor 594 (goat anti-mouse, 1:500, Abcam, Cambridge, UK).For other specific methods, reagents and instruments used, and parameter settings, please refer to our previously published literature.

| Data analysis
All experimental data are expressed as mean ± standard error values.Weekly comparison was made of defecation indicators among groups in Figure 1 using two-way ANOVA.Paired t-test was used for comparison before and after rotenone treatment or EA intervention, and an independent t-test was used for comparing two different groups.GraphPad Prism 9.4 (GraphPad Inc., La Holla, CA, USA) was used for data analysis with p < 0.05 considered to indicate statistical significance.

| Effect of EA on representative symptoms of PD constipation replicated in rotenone rat models
We assessed the success of model induction by measuring wet stool weight, ΔW, and stool count.After 28 days of rotenone treatment, rats in the PD group showed dry yellow-color stools (Figure 2), which was consistent with constipation.Compared with the SH group, rotenone injection significantly reduced the stool excretion (Figure 3a,g) and water content (Figure 3d) of the model rats.Consistent results were also obtained in the PD group before and after the rotenone intervention (Figure 3c,f,i).There was no difference in wet weight between the first week and the fourth week in the PD and EA groups.We considered that during this period the rats were in the growth phase and the amount of defecation increased accordingly with growth, thus leading to a non-significant reduction of wet weight.Therefore, we compared the wet weight of the PD group in the first and eighth weeks and found that the difference existed.
The stool properties of the PD group recovered after EA treatment (Figure 2).Compared to the PD group, EA treatment significantly increased stool excretion (Figure 3b,h) and water content (Figure 3e) in the EA group.The same results were seen in the EA group before and after EA treatment (Figure 3c,f,i).In conclusion, the EA group showed improvement in stool properties, excretion volume, and water content, suggesting the therapeutic effect of EA on PD constipation.

| Effect of EA on the autophagy pathway
To investigate the effect of EA on the expression of autophagy-related proteins, we measured the expression of ATG5 and LC3B in colonic tissues.Compared with the SH group, the levels of ATG5 (Figure 4a,c) and LC3II (Figure 4b,d) were significantly reduced in the colon tissues in the PD group, and their expression was upregulated by EA treatment.The expression level of LC3B mRNA could be corroborated with it (Figure 4e), but the ATG5 mRNA expression level did not change significantly.

| Effect of EA on the mitochondrial autophagy and elimination of α-Syn
To investigate the effect of EA on Parkin protein expression, we measured the expression of Parkin in colon tissue.Compared with the SH group, the level of Parkin significantly decreased in the colon in the PD group, and its expression was up-regulated by EA treatment (Figure 5).We further co-labeled Parkin with α-Syn by immunofluorescence (IF) to investigate whether EA was involved in α-Syn clearance through Parkin-induced mitochondrial autophagy.Compared with the SH group, the IF intensity of α-Syn in the colonic ENS increased in the PD group, suggesting α-Syn aggregation, and EA decreased α-Syn expression (Figure 6).IF double-labeling showed that Parkin and α-Syn were highly correlated in the colon (Figure 6a).

| DISCUSSION
Nowadays, there is a growing concern about the impact of non-motor PD symptoms on patients' quality of life.Constipation, as one of the most prominent non-motor symptoms of PD, is also a side effect of many PD drugs, such as anticholinergics and dopamine agonists in particular.PD constipation is the consequence of a combination of delayed colonic transit and dyssynergic defecation, but the effect of ENS injury on colonic motility is a key mechanism of PD constipation because, unlike colonic transit, dyssynergic defecation can be improved with apomorphine treatment or pelvic floor rehabilitation training (Fasano et al., 2015).Dopamine supplementation does not reduce PD constipation symptoms, which is consistent with colonic transit, suggesting that therapeutic studies on PD constipation should be centered on colonic transit.An in vitro study has demonstrated that rotenone-treated PD rats could exhibit a physiological defect of inhibitory neurons in the ENS, as evidenced by increased isometric contraction and decreased relaxation of the colonic longitudinal muscle (Greene et al., 2009).
In this study, a rotenone-induced model was recruited to characterize colonic dysfunction and ENS injury.The use of rotenone has been advantageous in modeling the pathogenesis of PD (Greenamyre et al., 2000).Rotenone, as a mitochondrial complex I inhibitor, can lead to selective degeneration of the nigrostriatal dopamine system through various pathogenic pathways, such as oxidative stress, α-Syn phosphorylation and aggregation, Lewy pathology, and proteasome dysfunction, which recapitulated the key pathological features of clinical PD (Betarbet et al., 2006).In contrast to the 1-Methyl-4-phenyl-1,-2,3,6-tetrahydropyridine (MPTP) model (Han et al., 2020), rotenone can reproduce the central and gastrointestinal features of PD in rats (Greene et al., 2009;Johnson et al., 2018), especially at low doses (2.5 mg/kg) of rotenone treatment, which has been further demonstrated in mice to provide replication of SNpc and myenteric nervous plexus neurodegeneration and the presence of α-Syn aggregates (Miyazaki et al., 2019).Rotenone reliably induced the PD model and had effects on the gastrointestinal tract, which contributed to the establishment of the PD constipation model.
As shown in Table 1, intestinal pro-secretory agents and gastrointestinal motility drugs are the main types of medications used clinically for PD constipation, which is consistent with chronic functional constipation, suggesting that the lubricating effect of intestinal secretions, in addition to colonic motility, is also one of the starting points for clinical treatment of PD constipation.Therefore, in the development of the model criteria, we not only observed stool properties (Figure 2) to visually determine constipation symptoms but also pinpointed the transit efficiency and secretory function of the colon by the stool excretion volume (Figure 3a,g) and water content (Figure 3d).We found that abnormalities in stool properties, excretion volume, and water content in the rotenone-induced rats, were consistent with the clinical features of constipation and suggested the successful establishment of the model.
EA, as a complementary and alternative therapy to treat a variety of diseases, refers to the application of a pulsating electrical current to stimulate acupoints through acupuncture needles (Zhang et al., 2020;Zhang & Li, 2021).Acupuncture has been shown to play a role in modulating the function of gastrointestinal motility (Li, 2015;Yin & Chen, 2010) and is widely used clinically.The effect of acupuncture at ST-25 on colonic motility has been demonstrated in clinical trials (Clinical Trials.govNCT01411501, NCT00508482) and animal models (Kim et al., 2006).Constipation has to do with the large intestine and is related to the spleen and stomach function.According to the principle of TCM, the large intestine front-mu acupoint ST25 is used to treat PD constipation.ST25 is the front-mu acupoint of the Yangming Stomach Meridian of Foot, where the Qi of the large intestine gathers.Modern medicine found that the sympathetic nerve of ST25 and the gastrointestinal tract are innervated by the same nerve segment.And ST25 is located in the abdomen with the small intestine in the deep of it, so it can provide good proximal treatment.
The neuronal protein α-Syn is thought to be central to the pathogenesis of PD.Excessive levels of wild-type α-Syn can lead to PD in familial selective cases, while accumulation of α-Syn protein occurs in sporadic PD (Xilouri et al., 2008).Natural α-Syn is usually monomeric, but the conformation seen in LBs is misfolded filaments that tend to aggregate.The oligomeric forms of α-Syn are now considered to be the most toxic, as they are thought to seed and accelerate the spread of α-Syn aggregates in the brain (Lama et al., 2021).Hilton et al. (2014) found that α-Syn deposition in the submucosal nerves of the stomach, duodenum, and colon of PD patients occurred in the prodromal phase of PD, indicating that pathological changes of ENS occurred in the early stage of PD and such pathological changes may migrate from ENS to the Central Nervous System (CNS) (Holmqvist et al., 2014), which suggested that the onset of PD may begin in the gastrointestinal tract and pathological changes in the ENS may be closely related to the onset of PD (Bagheri et al., 1999).IF results showed the presence of colon α-Syn aggregation in the PD group and EA treatment reduced its IF intensity (Figure 6), suggesting that EA could promote the clearance of abnormal α-Syn accumulation in the colon to repair the ENS and regulate colonic motility, which will improve the symptoms of PD constipation.The main modification of α-Syn in LB is the single phosphorylation of Ser-129 (Fujiwara et al., 2002).It has been proposed that α-Syn is heterodimerized after genetic mutation or environmental toxicity and phosphorylated to form strong neurotoxic pα-Syn, which is important progress in the study of PD pathogenesis in recent years (Grassi et al., 2018;Kam et al., 2018).The abnormal aggregation of pα-Syn occurs in both CNS and ENS, so the identification of clinical symptoms of ENS dysfunction and the detection of pα-Syn in the intestine may be important for the early diagnosis of PD (Yan et al., 2018).
Interestingly, we also found extensive α-Syn aggregation in the submucosal plexus of the colon in the PD group, suggesting that intestinal secretion may be involved in the formation of PD constipation.We did not further investigate the secretory motility and submucosal plexus of the colon, but our other study may be inextricably linked to it (the data are from another study that we have not yet fully completed and are used here for discussion purposes only).
Based on the 16s RNA [primers: 338F (ACTCCTAC GGGAGGCAGCAG) and 806R (GGACTACHVGGG TWTCTAAT)] results of sterile stools extracted from each group, we found that rotenone treatment increased the abundance and diversity of intestinal flora in the model rats (Figures 7a-c), suggesting that our model rats had a disturbed intestinal flora with increased numbers and proportions of harmful bacteria and that EA had a positive effect on this situation.A recent test in a mouse model found that mice overexpressing α-Syn showed reduced pathological signs after antibiotic treatment, also providing strong evidence for this idea (Sampson et al., 2016).β-diversity analysis showed significant differences in flora among the three groups (Figure 7d), suggesting a relationship between intestinal flora and PD constipation.This also seems to be associated with the factors of PD initiation.Proponents of the gut origin theory believe that intestinal flora affects PD pathophysiology through both neurological and humoral pathways.In short, it mediates α-Syn misfolding in the gut and brain through the inflammatory pathway (Hill et al., 2021).Although further studies are needed to determine their specific linkage, it is undeniable that the intestinal flora is inextricably linked to it.
Abnormal accumulation of neuronal α-Syn impairs the protein degradation system, especially the autophagylysosome pathway (Chen et al., 2015;Choubey et al., 2011;Huang et al., 2012).The inhibition of protein degradation in turn leads to the accumulation of α-Syn, thus creating a vicious cycle of pathological processes (Lin et al., 2019).Therefore, the way by which EA clears the abnormal accumulation of α-Syn in ENS is critical.There are three major pathways of protein degradation in eukaryotic cells, the autophagy-lysosome pathway, the proteasome pathway, and the caspase pathway, among which the caspase pathway is the pathway of protein degradation in apoptotic cells.The relatively tame degradation of α-Syn involves two major protein degradation pathways, the ubiquitindependent and -independent proteasome systems and the macroautophagy and CMA autophagy-endo-lysosomal systems (Grassi et al., 2018;LimanaQi et al., 2019;Rott et al., 2017).The ubiquitinated α-Syn in LB is mainly suffixed with one or two ubiquitin parts, with a very small fraction of three (Hasegawa et al., 2002;Tofaris et al., 2003).It is known that for ubiquitin-dependent degradation to occur, at least four ubiquitin splices on the protein are necessary (Pickart, 2000).Moreover, monoubiquitination of proteins cannot signal proteasomal degradation but may induce endocytosis or membrane transport (Hicke, 2001), which may involve α-Syn delivery between cells.Therefore, we hypothesized that the clearance of abnormal accumulation of α-Syn by EA is mainly achieved by the autophagic pathway.ATG5 is required for the formation of autophagic vesicles (Yousefi et al., 2006), and the expression level of ATG5 can roughly reflect the autophagic status of the organism.The results showed that EA increased the ATG5 level in the PD group (Figures 4a,c), suggesting that the autophagic pathway was indeed involved in the clearance of α-Syn in the EA group.We have made further tests on this basis.The formation of autophagosomes is essential for the development of autophagy, but only an intact autophagic stream can remove damaged mitochondria and abnormal proteins.LC3 is a signature protein in autophagosomal membranes and is divided into soluble LC3-I and membrane LC3-II.When autophagic signals are activated, a large number of autophagosomes are generated.LC3-II then binds to the autophagosome membrane, and the autophagic substrate protein Sequestosome-1 (SQSTM1/p62) enters the autophagosome via LC3-II leading to a decrease of its cytoplasmic level.When lysosomes bind to autophagosomes, the autophagosomes are degraded, and LC3-II concentration decreases (Tanida et al., 2008).That is, the number of autophagosomes was positively correlated with the LC3-II level and negatively correlated with the p62 level.The relative expression of LC3-II is often used to assess the level of autophagy.Our data clearly showed that EA upregulated LC3II protein and mRNA expression levels after rotenone treatment (Figure 4b,d,e), providing very strong evidence for our hypothesis.
Oligomeric α-Syn exerts its neurotoxic effects mainly through mitochondrial impairment, defective endoplasmic reticulum action, protease action, glial cell inflammatory response, cell membrane damage, lysosomal defects, and synaptic dysfunction (Zhang et al., 2018).In the neurodegenerative process of PD, mitochondrial damage, α-Syn aggregation, and protein degradation systems dysfunction interact with each other.α-Syn aggregation leads to a large number of mitochondrial defects, including reduced mitochondrial membrane potential and energy production, disruption of mitochondrialendoplasmic reticulum Ca2+ homeostasis, inhibition of mitochondrial dynamics, and induction of the mitochondrial release of the pro-apoptotic protein cytochrome c (Luth et al., 2014;Xie & Chung, 2012).When mitochondrial defects occur, mitochondrial-driven ROS production increases and triggers the input of α-Syn into the mitochondria, which induces α-Syn accumulation in the mitochondria and respiratory complex I dysfunction in turn (Ganguly et al., 2017;Perfeito et al., 2014).Overexpression or mutation of α-Syn impairs mitochondrial function, and the subsequent ROS increase leads to the toxicity of α-Syn conformation, thus creating a vicious cycle (Brundin et al., 2010;Pozo & Falzone, 2017).The accumulation of damaged mitochondria and α-Syn increases the proteasomal and autophagic lysosomal workload, which may lead to dysfunction of these clearance pathways in turn (Lee et al., 2010;Mercado et al., 2018).In sporadic Parkinson's disease (sPD), which accounts for 95% of Parkinson's disease cases, the autophagic mitochondrial pathway is blocked (Magalhaes et al., 2021), leading to a large accumulation of damaged mitochondria that are unable to generate sufficient energy for the cells, which in turn leads to gradual neuronal death and eventually to Parkinson's disease symptoms.Studies based on cadavers of Parkinson's patients strongly suggested the importance of mitochondrial damage in the pathogenesis of PD (Dexter et al., 1989).The E3 ubiquitin ligase Parkin plays an important role in the clearance of damaged mitochondria.Following mitochondrial membrane depolarization, Parkin translocates to the mitochondrial surface and mediates the degradation of defective mitochondria during mitochondrial autophagy (Hammerling et al., 2017;Narendra et al., 2008).Recent studies have found that Parkin overexpression plays a protective role against α-Syn-induced mitochondrial dysfunction (Wilkaniec et al., 2021).In this study, we found a significant increase in Parkin levels in the EA group (Figure 5a-c), which made us curious whether EA influenced the interrelationship between these two proteins, so we co-labeled these two proteins fluorescently in the same section.It turned out that Parkin and α-Syn in the colon were highly correlated (Figure 6a).Although there was no direct interaction, certainly, Parkin was involved in the clearance of mitochondria damaged by α-Syn.
In conclusion, our study demonstrated that EA could promote the clearance of α-Syn and damaged mitochondria by regulating colonic autophagy, thereby restoring the ENS function and acting as an ameliorator of PD constipation.However, there are several limitations of this study.First, because the autophagic pathway was not severed using methods such as gene knockdown or inhibitors, it can only be demonstrated that autophagy is involved in the regulation of PD constipation by EA.Whether autophagy is the main mechanism of EA intervention in PD constipation will be explored in future studies.Additionally, the necessity of a sham-operated group in EA studies remains controversial (Langevin et al., 2011), and the sham-EA group was not applied in this research.Therefore, the treatment of PD constipation may not entirely be mediated by EA in theory.In this study, conditions were not able to avoid the conduction of electricity to the colon tissue, so the effect of EA on ENS may partially be associated with direct stimulation of electric current.To more accurately describe the EA effect, we will include the sham group in further studies.

F
I G U R E 2 Representative stools of the SH, PD, and EA groups arranged in long diameters on graduated grid paper.EA, electroacupuncture; PD, Parkinson's disease