Andrographolide alleviates Parkinsonism in MPTP‐PD mice via targeting mitochondrial fission mediated by dynamin‐related protein 1

Abstract

Background and Purpose

Accumulating evidence indicates that mitochondrial dynamics play an important role in the progressive deterioration of dopaminergic neurons. Andrographolide has been found to exert neuroprotective effects in several models of neurological diseases. However, the mechanism of how andrographolide protects neurons in Parkinson's disease (PD) remains not fully understood.

Experimental Approach

Behavioural experiments were performed to examine the effect of andrographolide in 1‐Methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine (MPTP)‐PD mice. Mitochondrial mass and morphology were visualized using transmission electron microscopy (TEM). SH‐SY5Y cells and primary mouse neurons were exposed to rotenone to mimic PD in vitro. Western blotting, co‐immunoprecipitation and immunofluorescence were performed. The target protein of andrographolide was identified by biotin‐andrographolide pulldown assay as well as drug affinity responsive target stability (DARTS), cellular thermal shift (CETSA), and surface plasmon resonance (SPR) assays.

Key Results

Andrographolide administration improved behavioural deficits and attenuated loss of dopaminergic neurons in MPTP‐exposed mice and reduced cell death induced by rotenone in vitro. An increased mitochondrial mass, and decreased surface area were found in the striatum from MPTP‐PD mice, as well as in rotenone‐treated primary neurons and SH‐SY5Y cells, while andrographolide treatment preserved mitochondrial mass and morphology. Dynamin‐related protein 1 (DRP1) was identified as a target protein of andrographolide. Andrographolide bound to DRP1 and inhibited its GTPase activity, thereby preventing excessive mitochondria fission and neuronal damage in PD.

Conclusions and Implications

Our findings suggest that andrographolide may protect neurons against rotenone‐ or MPTP‐induced damage in vitro and in vivo through inhibiting mitochondrial fission.

What is already known

• Neuronal mitochondria with high levels of fission are often observed in PD patients and models

• Dynamin related protein‐1 (DRP1) is a key effector that mediates mitochondrial fission

What this study adds

• Andrographolide alleviates parkinsonism and mitochondrial dysfunction in PD models

• Andrographolide targets DRP1 to inhibit mitochondrial excessive fission in PD models

What is the clinical significance

• Andrographolide as a clinically used agent may represent a promising approach to treat PD.

• Blocking abnormal mitochondrial division and maintaining mitochondrial homeostasis may alleviate parkinsonism

Abbreviations

PD

Parkinson's disease

MPTP

methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine

TEM

transmission electron microscopy

DARTS

drug affinity responsive target stability assay

CETSA

cellular thermal shift assay

SPR

surface plasmon resonance assays

DRP1

dynamin‐related protein 1

T‐LA

the total time taken in the pole test

1. INTRODUCTION

Parkinson's disease (PD) is characterized by resting tremor, bradykinesia, and muscle rigidity in clinical manifestations and is the second most common neurodegenerative disorder, next to dementia, in the elderly (Pires et al., 2017). It is now well documented that aberrant mitochondria participate in the pathogenesis of PD. Mitochondria are not only double membraned organelles but also dynamic networks continuously undergoing biogenesis, fusion to join, fission to divide, and mitophagy to be cleared.

Recently, it has been reported that disturbed mitochondrial dynamics are central pathological components of PD (Bardai et al., 2018; Burte, Carelli, Chinnery, & Yu‐Wai‐Man, 2015; Wang, Wang, et al., 2016). Mitochondrial dynamics is primarily controlled by fission and fusion. Defects in mitochondrial dynamics would limit mitochondrial motility and lead to decreased energy production, excessive oxidative stress, and mtDNA deletion, finally resulting in cell death. Recently, the dynamin‐related protein 1 (DRP1) was identified as a contributor to the pathophysiology of PD (Rappold et al., 2014; Wu, Luo, & Tao, 2016). PD risk genes such as VPS35 mutations and α‐synuclein induce DRP1‐mediated fragmentation of mitochondria (Grassi et al., 2018; Martinez et al., 2018; Wang, Wang, et al., 2016). Under stress conditions, DRP1 is recruited from cytosol to the outer mitochondrial membrane and self‐assembled to form a spiral to tailor mitochondria, leading to mitochondrial depolarization and cell apoptosis due to the release of cytochrome c (Frank et al., 2001; Imoto, Tachibana, & Urrutia, 1998). A balance between fusion and fission is crucial to the survival of dopaminergic neurons, and DRP1 has been demonstrated to be a potential treatment target for PD (Wood, 2015). Thus, targeting DRP1 with low MW compounds may be a promising approach in PD.

https://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=9675 is a diterpenoid lactone isolated from Andrographis paniculata and has been shown to have anti‐inflammatory and antitumorigenic activities (Banerjee, Ahmed, Yang, Czinn, & Blanchard, 2016; Bao et al., 2009; Tran, Wong, & Chai, 2017; Yang et al., 2017; Zhou et al., 2012). In previous studies, we demonstrated that andrographolide inhibited https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1770, as well as the Th1/Th17 response in colitis (Guo et al., 2014; Liu et al., 2014), stabilized https://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=910 protein to reverse resistance to https://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=4789 in colorectal cancer (Wang, Guo, et al., 2016), inhibited ROS‐mediated NF‐κB activation (Peng, Gao, et al., 2016), and down‐regulated MAPK and NF‐κB pathways to ameliorate lung injury (Peng, Hang, et al., 2016). Andrographolide can cross the blood–brain barrier and concentrate in the brain (Li et al., 2012; Lu, 1995), where it ameliorated oxidative injury through suppressing https://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=673/https://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=285 and NF‐κB pathways to protect neurons in ischaemic stroke models (Chan, Wong, Wong, & Bian, 2010; Chern et al., 2011), and promoted hippocampal https://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=4872 signalling cascade to reverse depressant‐like effects (Zhang et al., 2019). It can also attenuate cognitive impairment in the mouse model of Alzheimer's disease (Geng et al., 2018; Rivera et al., 2016; Serrano et al., 2014). In addition, andrographolide and its analogue inhibited microglia activation to reduce dopaminergic neurodegeneration (Wang, Liu, Zhang, Wilson, & Hong, 2004; Zhang et al., 2014). However, whether andrographolide protects mitochondria in PD remains unknown.

Here, we tested if andrographolide protected dopaminergic neurons through preserving mitochondrial functions in MPTP‐mice and in rotenone‐treated neurons and SH‐SY5Y cells. In particular, we focused on DRP1‐mediated mitochondrial fission as the regulatory mechanism of andrographolide. Our key finding showed that andrographolide inhibited the GTPase activity of DRP1 through directly binding to DRP1, thus interrupting DRP1‐mediated mitochondrial fission and preserving mitochondrial functions to protect against rotenone‐induced injury. Taken together, our results demonstrated that andrographolide, a clinically used agent, can serve as a promising approach to treat PD through blocking abnormal mitochondrial division and maintaining mitochondrial homeostasis.

2. METHODS

2.1. Cell line and mesencephalic primary neuron culture

SH‐SY5Y human neuroblastoma cells (ATCC Cat# CRL‐2266, RRID:CVCL_0019) were purchased from ATCC and cultured in RPMI 1640 medium (Gibco, Herndon, USA) containing 10% FBS (Gibco, Herndon, USA) in a humidified incubator at 37°C and 5% CO₂.

Ventral mesencephalon was dissected from E12.5 C57BL/6 embryos in cold 0.01‐M PBS on ice, and DMEM containing 0.25% trypsin was added and incubated for 10 min in a 37°C water bath. The tissue pieces were dissociated by a 1‐ml pipette tip through pipetting five to seven times. The enzymatic digestion was terminated by adding DMEM/F12 (Gibco) supplemented with 10% FBS. The cells were pelleted by centrifugation for 3 min at 500× g at room temperature, seeded on 96‐well plate coated with poly‐d‐lysine, and cultured in B27‐supplemented neurobasal medium (Gibco, A3582801 and A3582901). Half of the media was replaced with fresh media every other day until the seventh day.

2.2. Cell viability assay

The cells were plated at a density of 4 × 10³ viable cells per well in 96‐well plates. Cell were treated with 0.1‐, 0.3‐, and 1‐μM andrographolide for 3 hr, followed by 30‐μM rotenone for 6 hr. Then cell viability was measured by MTT assay. Annexin V/PI staining was performed according to manual and analysed by BD FACSCalibur™ flow cytometer (BD Biosciences). Annexin V+/PI− and annexin V+/PI+ cells were considered as apoptotic cells in the early and late phases.

2.3. ROS, mitochondria membrane potential analysis, and ATP content

Intracellular ROS was determined using DCFH‐DA stain and mitochondrial membrane potential was measured by staining with JC‐1. After treatment with andrographolide and rotenone, cells were washed with 0.01‐M PBS and incubated with 20‐μM DCFH‐DA or 5 μg·ml⁻¹ JC‐1 at 37°C for 30 min, and then cells were visualized using an inverted fluorescence microscope (Olympus, Japan, 20× objective) or measured with fluorescence spectrometry (Spectra MaxGemini, Molecular Devices Corporation, USA). The ATP content was assessed by the luciferin–luciferase method according to the manual. The activities of MDA (A003‐1), SOD (A001‐2‐1), and catalase (A007‐1‐1) were detected using reagent kit (Jiancheng, Nanjing, Jiangsu, China).

2.4. Isolation of mitochondria

Cells were harvested by centrifugation at 600× g for 10 min, washed with PBS, and resuspended with five volumes of Solution A (0.25‐M sucrose, 20‐mM HEPES‐KOH, pH 7.5, 10‐mM KCl, 1.5‐mM MgCl₂, 1‐mM EDTA, 1‐mM EGTA, 1‐mM DTT, and 0.1‐mM PMSF). The cellular suspension was homogenized with a Teflon glass homogenizer on ice. The homogenate was then centrifuged at 750× g for 10 min. The resulting supernatant was collected and then centrifuged at 10,000× g for 15 min. The pellet contained the mitochondria.

2.5. Oxygen consumption

As described by Zhang, Khvorostov, et al. (2011), a Clark‐type O₂ electrode was used in a magnetically stirred and thermostatically controlled 0.3‐ml chamber at 25°C (Oxytherm; Hansatech). Cells (5 × 10⁶) in 2‐ml buffers (0.137‐M NaCl, 5‐mM KCl, 0.7‐mM NaH₂PO₄, 25‐mM Tris–HCl, pH 7.4) were rated for O₂ consumption.

2.6. Western blot analysis

The cells were washed three times with ice‐cold PBS and then lysed in lysis buffer (P0013, Beyotime, Nantong, China) for 30 min on ice. Protein lysate was centrifuged at  10,000× g (4°C) for 15 min and denatured by boiling with loading buffer. Samples containing 30‐μg proteins were separated by 12.5% SDS‐PAGE and transferred onto PVDF membranes. After being blocked with 5% non‐fat milk at room temperature for 1 hr, membranes were incubated with primary antibodies (1:500) at 4°C overnight and then incubated with HRP‐conjugated secondary antibodies (1: 5,000) for 2 hr at room temperature. Detection was carried out using the super enhanced chemiluminescent (ECL) detection kit (Millipore, USA). Densitometry results were normalized by β‐actin using Image J software (version 1.37 for Windows; National Institutes of Health, Bethesda, MD, USA, RRID:SCR_003070).

2.7. Co‐IP

After andrographolide and rotenone treatment, proteins were extracted using cold RIPA followed by centrifuging at 14,000× g 4°C for 15 min. One microgram of primary antibody against DRP1 and 20‐μl protein A/G plus agarose were added into the supernatant at 4°C overnight. After centrifugation at 500× g for 3 min, the pellet was washed three times with pre‐cooled PBS , and the beads were boiled in 2× loading buffer (100‐mM Tris–HCl [pH 6.8], 4% SDS, 1% bromophenol blue, 20% glycerol, and 2% β‐mercaptoethanol). Then the supernatants were collected and subject to western blot analysis for DRP1, FIS1, MFF, MiD49, and MiD51.

2.8. Cross‐linking

Dimeric DRP1 was cross‐linked with a membrane‐permeable linker (EGS, Pierce). After treatment, SH‐SY5Y cells were incubated with 2‐mM EGS for 45 min at room temperature and then quenched in 20‐mM Tris–HCl for 15 min. After a rinse with ice‐cold PBS, SH‐SY5Y cells were lysed, and DRP1 was detected using western blot analysis.

2.9. Immunofluorescence

The immuno‐related procedures used comply with the recommendations made by the British Journal of Pharmacology (Alexander et al., 2018). Cells were cultured with MitoTracker Green for 30 min, washed three times with 0.01‐M PBS, fixed with 4% paraformaldehyde for 30 min, and permeabilized with 0.3% Triton X‐100 for 30 min. After blocking with 3% BSA for 1 hr, cells were incubated with antibodies against DRP1 at room temperature for 2 hr. Slides were washed three times with PBS and incubated with Alexa Fluor 594‐conjugated secondary antibodies (1:200) for 1 hr at room temperature. Nuclei were stained with DAPI (10 μg·ml⁻¹) for 3 min. Images were acquired using confocal microscopy (LSM710, Carl Zeiss, Germany).

2.10. Transmission electron microscopy

SH‐SY5Y cells were harvested using cell scraper and fixed with ice‐cold 2.5% glutaraldehyde. Striatum from mice were also fixed in 2.5% glutaraldehyde and then fixed in OsO4 and sliced into 1‐μm sections. The ultrastructure and the number of mitochondria were observed per section. All the evaluators were blinded to the groups. As described by Ding, Zhao, Li, Gong, and Lu (2015) and Liang et al. (2010), the Image J analysis software was used in stereology to calculate stereological mitochondrial parameters such as the number of mitochondria per picture, mean surface area (S), surface‐to‐volume ratio (Rsv), and number density (Nv).

2.11. RNA interference

siRNA duplexes for DRP1 and negative control were designed by GenePharma (Shanghai, China). Adhered SH‐SY5Y cells at 50% confluence were transfected with siRNA of DRP1 (5′‐GAGGUUAUUGAACGACUCA‐3′) and negative control (5′‐UUCUCCGAACGUGUCACGU‐3′) for 48 hr using lipofectamine 2000 (Invitrogen, Carlsbad, CA), according to the manufacturer's instructions.

2.12. Pulldown of andrographolide‐bound proteins

SH‐SY5Y cell lysates were incubated with 10‐μM biotin, 10‐μM biotin‐andrographolide, and 10‐μM biotin‐andrographolide plus 100‐μM andrographolide overnight at 4°C, respectively. The lysates were pulled down with streptavidin‐conjugated beads (Softlink soft release Avidin Resin, PROMEGA) at 4°C for another 4 hr. After an extensive wash with PBS, the beads were boiled in 2× loading buffer. Then the supernatants were collected and subjected to western blotting for DRP1.

2.13. Cellular thermal shift assay (CETSA)

SH‐SY5Y cells were incubated with or without andrographolide (1 μM) for 9 hr, and then the cells were collected and subjected to CETSA assay (Jafari et al., 2014). Briefly, incubated cells were equally divided into 10 parts, each part was heated for 3 min under different temperature (43, 46, 49, 52, 55, 58, 61, 64, 67, and 70°C), and then the heated cells were kept at −80°C for 12 hr, then at room temperature for 5 min, and the process repeated one more time. After that, cell lysates were extracted by centrifugation at 20,000× g, 20 min. Levels of DRP1 were assessed by western blot.

2.14. Drug affinity responsive target stability (DARTS)

Lysates from SH‐SY5Y cells were incubated with or without andrographolide (1 μM) for 9 hr at room temperature as described by Lomenick et al. (2009). Lysates were then divided into seven portions for digestion with different concentrations of pronase E (1074330001, Sigma Aldrich) for 30 min at room temperature. Samples were boiled immediately after adding loading buffer to stop the digestion. Each sample was loaded onto 15% SDS‐PAGE for western blotting.

2.15. Surface plasmon resonance (SPR)

SPR assays was performed using the Biacore T200 as follows. Recombinant human DRP1 protein (ab153041, Abcam) was immobilized on a Biacore CM5 sensor chip via the primary amine groups. The compounds were superfused at a rate of 30 μl·min⁻¹ for 60 s to allow for association, followed by 150 s for dissociation over immobilized protein in PBS/5% DMSO running buffer (1.05× PBS, 0.5% P20 surfactant, 5% DMSO, pH 7.4). Andrographolide was tested for binding at 1.5625 to 50 μM. Normalization of the data involved transformation of the y‐axis such that the theoretical maximum amount of binding for a 1:1 interaction with the protein surface corresponded to a sensor response of 100 relative units (RU).

2.16. GTPase activity of DRP1

After treatment, cells were washed three times with iced PBS and lysed with lysis buffer. Twenty microlitres of GTP‐agarose (G9768; Sigma) were added to cell lysates at 4°C overnight. The beads were collected (500× g for 3 min) and washed with iced PBS for three times and then boiled in 2× loading buffer. DRP1 bound to the agarose was detected using western blot.

The GTPase activity of DRP1‐GTPase was measured using GTPase Activity Assay Kit (MAK113, Sigma‐Aldrich) in the following buffer: 50‐mM NaOH, 10‐mM MgCl₂, 200‐mM KCl, pH 6.5. The reaction was initiated with the addition of GTP and 200 μg·ml⁻¹ DRP1‐GTPase, followed by incubation at 30°C for indicated times. GTP hydrolysis was measured by the amount of released Pi.

2.17. Mice and MPTP‐PD model

Animal welfare and experimental procedures were carried out strictly in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, USA) and the related ethical regulations of our university. All efforts were made to minimize animals' suffering and to reduce the number of animals used. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny, Browne, Cuthill, Emerson, & Altman, 2010; McGrath, Drummond, McLachlan, Kilkenny, & Wainwright, 2010) and with the recommendations made by the British Journal of Pharmacology. C57BL/6 mice (male, 10–12 weeks old, 23–25 g) were purchased from Model Animal Genetics Research Center of Nanjing University (Nanjing, China). The animals were housed, five per cage with food and water ad libitum, on a 12‐hr light/dark cycle with lights on at 06:00 hr and controlled (22–23°C) temperature.

The MPTP‐PD model was prepared as described previously (Jackson‐Lewis & Przedborski, 2007). Adult male C57BL/6 mice were randomly divided into five groups (10 mice per group). The control group was treated with PBS (vehicle). Mice were injected intraperitoneally with MPTP (dissolved in PBS) in a final concentration of 25 mg·kg⁻¹ daily for five consecutive days. One hour after MPTP injection, mice were given 2.5 or 5 mg·kg⁻¹ andrographolide intraperitoneally once a day for 12 days. Levodopa was orally administered (75 mg·kg⁻¹) as a positive control. After the final treatment, behavioural procedures including pole test, beam hang test, rotarod task, and open field and forced swimming tests were conducted to assess behavioural changes. All the evaluators were blinded to the treatment groups.

2.18. Pole test

The pole test is conducted as previously described (Luchtman, Shao, & Song, 2009). A 1 cm in diameter and 45 cm in height wooden pole was entangled with gauze and vertically fixed in the cage. A board was fixed horizontally to avoid mice sitting on the top of pole. The time that mice spent to turnover head (T turn) and reach to the bottom of cage (T‐LA) was measured. Each animal performs three successive training before MPTP administration. After the end of andrographolide treatment, the test is repeated for three trials per mouse, and the average T turn and T‐LA per mouse are used to assess the motor coordination and grasping capability.

2.19. Beam hang test

The protocol was adapted from Masuda‐Suzukake et al. (2014). The equipment is composed of a round wooden beam (diameter: 0.5 cm) to allow the mice to grasp it and a smooth lid to prevent mice sitting on the beam. Mice were trained by hanging on it successfully during 120 s before MPTP administration. After the end of andrographolide treatment, the time mice spent hanging on the beam was recorded to assess the motor coordination and grasping capability.

2.20. Rotarod task

The rotarod task was performed as described (Rahimmi, Khosrobakhsh, Izadpanah, Moloudi, & Hassanzadeh, 2015). Mice were positioned on the rotarod consisting of a suspended rod (diameter: 4 cm), which was programmed to increase speed gradually from 1 to 20 rpm in 180 s. Mice were trained over three consecutive days, and at each day of training, they were placed on the rod for a session of five trials. The time taken for the mouse to fall from the rotator is a measure of motor coordination.

2.21. Immunohistochemical analysis, TUNEL, and Nissl staining

Immunohistochemical analysis was performed on paraffin‐embedded brain tissue sections (3 μm). Briefly, the sections were deparaffinised, rehydrated, and washed in 1× PBS and then treated with 2% hydrogen peroxide, blocked with 3% goat serum, and incubated with anti‐TH (1:100) overnight at 4°C. The slides were then processed with GTVisin™ anti‐mouse/anti‐rabbit immunohistochemical analysis KIT according to the manufacturer's instructions.

TUNEL and Nissl staining were performed to examine apoptosis and neuron damage, respectively, according to the manual.

2.22. Data and statistical analysis

The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology. Data are expressed as mean ± SEM. One‐way ANOVA test with Tukey post hoc test and two‐way ANOVA test were used for statistical evaluation. All statistical analyses were conducted using GraphPad Prism Software Version 7.0 (GraphPad Software Inc., La Jolla, CA, RRID:SCR_002798). P values of <.05 were considered statistically significant.

2.23. Materials

Primary antibodies against DRP1(sc‐32898, RRID:AB_2093533), MFN2(sc‐50331, RRID:AB_2142754), OPA1(sc‐30572, RRID:AB_2158165), FIS1(sc‐98900, RRID:AB_2246809), MFF (sc‐398731), protein A/G plus‐agarose (sc‐2003, RRID:AB_10201400), TH (sc‐14007, RRID:AB_671397), and actin (sc‐8432, RRID:AB_626630) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Primary antibodies against DAT (22524‐1‐AP), MAO‐B (12602‐1‐AP, RRID:AB_2137273), MiD49 (16413‐1‐AP, RRID:AB_2714217), MiD51 (20164‐1‐AP, RRID:AB_10639522), and Alexa Fluor 594‐conjugated donkey anti‐mouse secondary antibody (SA00006‐7) and Alexa Fluor 488‐conjugated donkey anti‐rabbit secondary antibody (SA00006‐6) were purchased from Proteintech Group (Chicago, USA). COX IV (#4844, RRID:AB_2085427), CytoC (#4272, RRID:AB_2090454), Bcl2 (#3498, RRID:AB_1903907), Bax (#2772, RRID:AB_10695870), and https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1619 (#9662, RRID:AB_331439) were procured from Cell Signaling Technology (Beverly, MA, USA). JC‐1 dye (C2005), DCFH‐DA (S0033), MTT (ST316), annexin V/PI staining kit (C1062), ATP detection kit (S0026), and Nissl staining kit (C0117) were bought from Beyotime Company (Nantong, China). GTVisin™ anti‐mouse/anti‐rabbit immunohistochemical analysis kit (GK500705) was purchased from Gene Company (Shanghai, China). TUNEL analysis KIT (A112‐01) was purchased from Vazyme Company (Nanjing, China). Rotenone (Sigma‐Aldrich, MO, USA) stock in DMSO was prepared fresh for every experiment. All other chemicals were obtained from Sigma‐Aldrich.

2.24. Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 (Alexander, Fabbro et al., 2017a, b; Alexander. Kelly et al., 2017).

3. RESULTS

3.1. Andrographolide attenuated PD‐like phenotypes and maintained mitochondrial number in MPTP‐induced PD

To assess the protective effects of andrographolide on PD, an MPTP‐induced PD animal model was employed. Andrographolide or levodopa were given as shownin Figure 1a, and a series of behavioural tests were performed.

Figure 1.

Andrographolide (Andro) improved motor deficits, reduced neuronal loss and oxidative stress, and preserved mitochondrial morphology in MPTP‐challenged mice. (a) Overview of the experimental design. (b) The pole test was used to measure the motor function. The time from when the mice were positioned head‐upward near the top of the pole until they turned completely downward was recorded and indicated as T turn, while the total time taken in the pole test was indicated as T‐LA. (c) Beam hang test was used to measure the grip strength and motor coordination. (d) The effect of andrographolide on the fall latency of rotarod test in the MPTP‐induced PD mouse model. (e) The expression of TH, the dopamine transporter (DAT), and MAO‐B in the striatum of each group were measured using western blot analysis. TH, DAT, and MAO‐B protein were normalized by actin. (f) Immunohistochemical staining of TH‐positive cells in the substantia nigra. (g) Malondialdehyde (MDA) production, SOD activity, and catalase activity among striatum were determined as the indicators of oxidative stress. (h) The mitochondrial morphology in each group was assessed by transmission electron microscope (TEM). Data represent mean ± SEM, n = 10 in b, c, and d, n = 5 in e and g. #P < .05, significantly different from control group; *P < .05, significantly different from MPTP group

The pole test is utilized to examine motor coordination. Data reported in Figure 1b indicated that MPTP‐PD mice have poor coordinative ability with a longer time spent turning round (T turn) and finishing total distance (T‐LA). Coordination was obviously improved with treatment of 5 mg·kg⁻¹ andrographolide or levodopa. The beam hang test was performed to assess muscle tone via measuring grasp capability. Similarly, in the beam hang test, MPTP‐PD mice exhibited poor grasping capability with a shorter time to fall down, and 5 mg·kg⁻¹ andrographolide treatment improved grasping ability in MPTP‐PD mice (Figure 1c). The rotarod task was utilized to evaluate movement coordination. As described in Figure 1d, mice exposed to MPTP exhibited poor movement coordination along with a shorter time spent on the rod compared with normal mice, and 5 mg·kg⁻¹ andrographolide significantly increased the time spent on the rotarod. Depressive and anxiety disorders are often comorbid with PD (Schapira, Chaudhuri, & Jenner, 2017). Thus, an open field experiment and a forced swimming test were performed to measure depressive symptoms in mice. The distance travelled and time spent in the central area by MPTP‐PD mice was significantly reduced compared to the control group. In this experiment, 5 mg·kg⁻¹ andrographolide treatment increased total distance moved and time spent in central area (Figure S1A). As shown in Figure S1B, MPTP‐PD mice also exhibited shorter swimming time and stayed afloat longer than normal mice. After 5 mg·kg⁻¹ andrographolide treatment, the swimming time was markedly increased, indicating improved depressive symptoms.

The results from the behavioural experiments demonstrated that andrographolide can improve MPTP‐induced behavioural impairment. MPTP neurotoxicity is dependent on its conversion by MAO‐B to N‐methyl‐4‐phenylpyr‐idine (MPP⁺). To further confirm the protective effects of andrographolide on dopaminergic neurons, we measured expression of https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1243, a marker of dopaminergic neurons, the dopamine transporter, https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=927 and https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2490 in the striatum using western blot analysis. We also detected TH expression in the substantia nigra using immunohistochemistry. The levels of TH and DAT were decreased in MPTP‐PD mice, while andrographolide significantly elevated TH and DAT levels (Figure 1e,f). The expression of MAO‐B did not change among groups, suggesting andrographolide had no effect on the metabolism of MPTP.

MPP⁺, an inhibitor of mitochondrial electron transport chain complex I, causes oxidative stress and apoptosis in dopaminergic neurons. The content of malondialdehyde (MDA) and the activity of antioxidant enzymes such as SOD and catalase were examined to assess the extent of oxidative stress (Figure 1g). In this experiment, 5 mg·kg⁻¹ andrographolide minimized MDA content and elevated activity of SOD and catalase. Apoptosis and neuron injury in the striatum measured by TUNEL and Nissl staining were induced by MPTP, which was inhibited by 5 mg·kg⁻¹ andrographolide treatment (Figure S1C,D).

As mitochondria play a key role in oxidative stress and apoptosis, we observed the morphology of mitochondria in the striatum, using transmission electron microscopy (TEM; Figures 1h and S1E). Considerably more small mitochondria were found in the MPTP group than in normal mice. Mitochondria from 5 mg·kg⁻¹ andrographolide‐treated mice exhibited elongated appearance, and the number of mitochondria per area was significantly lower compared to the MPTP group.

Taken together, these data provided evidence that andrographolide attenuated behavioural impairment, dopaminergic neuron loss, oxidative stress, and preserved mitochondrial morphology in mice with MPTP‐induced PD.

3.2. Andrographolide remedied mitochondrial dysfunction in rotenone‐exposed neurons and SH‐SY5Y cells

To examine whether andrographolide has the capacity to protect mitochondria directly in neurons, rotenone, a potent inhibitor of the mitochondrial complex I, was utilized to mimic the pathological conditions of mitochondria of PD in both primary neurons and SH‐SY5Y cells. Andrographolide (0.1, 0.3, and 1 μM) alone has no discriminative cytotoxic effect on SH‐SY5Y cells (Figure S2A). As shown in Figure 2a,b, andrographolide (0.1, 0.3, and 1 μM) was cocultured with primary neurons for 3 hr followed by rotenone (30 μM) for 6 hr, the morphology was photographed, and cell viability was assessed by MTT assay. Rotenone 30 μM down‐regulated cellular viability and 1‐μM andrographolide increased cell viability in primary neurons, as well as in rotenone‐treated SH‐SY5Y cells. The morphology image also confirmed these protective effects.

Figure 2.

Andrographolide (Andro) reduced rotenone‐induced toxicity and attenuated mitochondrial dysfunctions. (a) Primary neurons or (b) SH‐SY5Y cells were pretreated with andrographolide (0.1, 0.3, and 1 μM) for 3 hr followed by 30‐μM rotenone for 6 hr. The morphological changes were pictured, and cellular viability was measured with MTT assay. (c, d) ROS generation and mitochondrial membrane potential of SH‐SY5Y cells after treatment were assessed using DCFHDA staining and JC‐1 staining. Scale bar: 20 μm. (e, f) The oxygen consumption and ATP production of SH‐SY5Y cells after treatment were measured. (g) After andrographolide pretreatment, mitochondria in rotenone‐exposed SH‐SY5Y cells were isolated, and the release of cyto c from mitochondria to cytoplasm was detected using immunoblotting. Cyto c protein was normalized with COX IV (mitochondria) and actin (cytoplasm). Data represent mean ± SEM, n = 5. #P < .05, significantly different from control group; *P < .05, significantly different from rotenone group

Mitochondrial dysfunction affects oxidative stress and oxidative phosphorylation. Therefore, we next measured ROS level and mitochondrial membrane potential using DCFH‐DA and JC‐1 staining. Results in Figure 2c,d showed that andrographolide attenuated rotenone‐induced intracellular ROS levels and mitochondrial membrane potential collapse. Andrographolide also decreased MDA content and restored the activity of SOD, indicating that oxidative stress in rotenone‐exposed SH‐SY5Y cells was attenuated by andrographolide (Figures S2B and 2c).

We further measured whether mitochondrial dysfunction due to rotenone exposure can be reversed by andrographolide. An examination of the O₂ consumption rate (OCR) showed that the rotenone‐treated group consumed O₂ at 2 nmol·min⁻¹/1 × 10⁴ cells (Figure 2e). In contrast, the group pretreated with 1‐μM andrographolide consumed O₂ at 2.7 nmol·min⁻¹/1 × 10⁴ cells, which was significantly higher than that of rotenone alone. ATP produced from oxidative phosphorylation is primarily completed in mitochondria. ATP level was decreased after rotenone treatment, which was reversed by andrographolide in a dose‐dependent manner (Figure 2f).

Mitochondrial dysfunction can also lead to apoptosis through releasing cytochrome c (cyto c). As depicted in Figure 2g, andrographolide reversed rotenone‐induced release of cyto c from mitochondria in a dose‐dependent manner. Andrographolide also inhibited rotenone‐stimulated cleaved caspase‐3 and apoptosis (Figures S2D and 2e).

In conclusion, andrographolide attenuated oxidative stress, preserved membrane potential and oxidative phosphorylation, and inhibited cyto c‐mediated apoptosis in rotenone‐treated cells.

3.3. Andrographolide preserved mitochondrial morphology and number in rotenone‐treated SH‐SY5Y cells

On the basis of the finding that andrographolide decreased abnormal fragmented mitochondria in the MPTP‐PD model and preserved mitochondrial function to counter rotenone‐triggered damage in vitro, we investigated whether andrographolide can maintain mitochondrial morphology under rotenone‐exposed conditions. We first observed the mitochondrial morphology in neurons using MitoTracker Green. As exhibited in Figure 3a, rotenone triggered smaller and more dispersed mitochondria with weak fluorescence in neurons, while andrographolide pretreatment preserved tight mitochondria distribution with stronger fluorescence compared with rotenone treatment. Moreover, mitochondria exhibited a smaller and fractured morphology in rotenone‐treated SH‐SY5Y cells, and andrographolide treatment preserved elongated mitochondrial networks (Figure 3b).

Figure 3.

Andrographolide (Andro) preserved mitochondrial morphology and number in rotenone‐treated SH‐SY5Y cells. (a) After being cultured at 0.1‐, 0.3‐, and 1‐μM andrographolide for 3 hr, followed by the addition of 30‐μM rotenone for 6 hr, the mitochondrial morphology in neurons was pictured using MitoTracker Green. Scale bar: 10 μm. (b) After being cultured with 0.1‐, 0.3‐, and 1‐μM andrographolide for 3 hr, followed by the addition of 30‐μM rotenone for 6 hr, the mitochondrial morphology and distribution in SH‐SY5Y cells were pictured using immunofluorescence against TOMM20. Scale bar: 10 μm. (c) The mitochondrial morphology in SH‐SY5Y cells was assessed by transmission electron microscope. The (d) mitochondrial mean surface area, (e) surface‐to‐volume ratio, and (f) mitochondrial number density were calculated as described in Section 2. Data are mean ± SEM, n = 9, in d, e, and f. #P < .05, significantly different from control group; *P < .05, significantly different from rotenone group

Next, transmission electron microscopy (TEM) was further employed to examine mitochondrial morphology in SH‐SY5Y cells. As shown in Figure 3c, andrographolide can prevent mitochondria being compacted under rotenone‐exposed condition. Mdivi‐1, an inhibitor of mitochondrial division, was used to provide a positive control group. Furthermore, cells treated with andrographolide alone showed no obvious change in the mitochondrial network. Stereological analysis of mitochondrial number density (Nv), mitochondrial surface‐to‐volume ratio (Rsv), and mitochondrial mean surface area (S) also confirmed that andrographolide preserved mitochondrial morphology and network (Figure 3d–f).

These results illustrated that andrographolide preserved mitochondrial morphology against rotenone‐stimulated damage.

3.4. Andrographolide inhibited mitochondrial translocation of DRP1

Mitochondrial morphology and number are controlled dynamically by mitochondria biogenesis, mitophagy, and mitochondrial fission–fusion. Mitochondrial biogenesis indicated by transcription factors such as POLG1, POLG2, SSBP1, TFAM, and TWINKLE and autophagy monitored by LC3 were examined. As shown in Figure S3A, rotenone decreased mitochondrial biogenesis and increased autophagy. However, andrographolide improved mitochondrial biogenesis and inhibited rotenone‐induced autophagy.

As andrographolide treatment decreased the number of smaller mitochondria in vivo and in vitro, we wondered whether the process of mitochondrial fission and fusion was affected by andrographolide. DRP1, which is responsible for mitochondrial fission, was translocated to mitochondria after rotenone exposure. Both andrographolide and Mdivi‐1 pretreatment decreased the translocation of DRP1 (Figure 4a). Moreover, rotenone‐induced mitochondrial fission and andrographolide‐mediated protection were not related to mitochondrial fusion, which was demonstrated by a lack of change in expression as well as location of OPA1 and MFN2 (Figure 4a). We also examined the co‐localization between DRP1 and mitochondria through immunofluorescence. As shown in Figure 4b, rotenone treatment promoted translocation of DRP1 to mitochondria while andrographolide diminished the colocalization between DRP1 and mitochondria. Andrographolide also preserved elongated mitochondrial morphology. DRP1 oligomer complexes were further ascertained using EGS crosslinking assay. As exhibited in Figure 4c, rotenone treatment caused DRP1 oligomerization to scissor mitochondria, while andrographolide decreased its oligomerization. DRP1 cooperates with FIS1, MFF, MiD49, and MiD51 to carry out mitochondrial fission. Co‐IP and immunofluorescence demonstrated that andrographolide decreased the interaction between DRP1 and FIS1 as well as MFF (Figure 5a,b). These results indicated that andrographolide hampered DRP1‐mediated mitochondrial fission triggered by rotenone.

Figure 4.

Andrographolide (Andro) inhibited mitochondrial translocation and oligomerization of DRP1. (a) Pretreatment with 1‐μM andrographolid or 20‐μM Mdivi‐1 for 3 hr followed by rotenone for 6 hr; the cytoplasm and mitochondria were separated, and the translocation of DRP1 between cytoplasm and mitochondria were detected using immunoblotting. Anti‐actin in lysates was performed to normalize for cytoplasmic proteins while anti‐COX IV was to normalize for mitochondrial proteins. (b) The DRP1 distribution was detected with immunofluorescence using MitoTracker Red and anti‐DRP1 (green) after andrographolide and rotenone treatment. (c) Pretreatment andrographolide at 1 μM for 3 hr followed rotenone for 6 hr; EGS was added to cross‐link interacted proteins. The oligomers of DRP1 were assessed using western blot analysis. Data are mean ± SEM, n = 5. #P < .05, significantly different from control group; *P < .05, significantly different from rotenone group

Figure 5.

Andrographolide (Andro) reduced colocalization between DRP1 and FIS1, DRP1, and MFF. (a, b) SH‐SY5Y cells were precultured with 1‐μM andrographolide for 3 hr followed by 30‐μM rotenone for 6 hr, and the interactions between DRP1 and MFF, DRP1 and FIS1, DRP1 and MiD49, and DRP1 and MiD51 were detected by co‐immunoprecipitation and immunofluorescence. FIS1 and MFF were normalized with DRP1. Data are mean ± SEM, n = 5. #P < .05, significantly different from control group; *P < .05, significantly different from rotenone group

3.5. Andrographolide directly bound to DRP1 and inhibited its GTPase activity to protect against rotenone‐induced damage

As andrographolide markedly restrained the mitochondrial translocation and oligomerization of DRP1, we wondered if there was a direct interaction between andrographolide and DRP1. Biotin‐andrographolide was synthesized and used to pull down the target proteins. As shown in Figure 6a, biotin‐andrographolide bound to DRP1, which competed with unlabelled andrographolide. When drug binds to its target protein, the protein will be resistant to enzymic degradation or denature at high temperatures. Therefore, CETSA and DARTS were used to further confirm this direct interaction. After exposure to 1‐μM andrographolide for 9 hr, the level of DRP1 was higher than in the untreated group under 0.01% pronase E condition. This effect was also shown to be dose‐dependent (Figure 6b). DRP1 started to degrade at 49°C and almost disappeared at 58°C in vehicle‐treated cells, while it degraded at 55°C and disappeared at 61°C in andrographolide‐treated cells (Figure 6c). Next, we treated cells with three doses of andrographolide followed by heating at 52°C. The results showed andrographolide increased stability of DRP1 in a dose‐dependent manner (Figure 6c). To quantify the interaction between andrographolide and DRP1, SPR by Biacore 200 was employed. The affinity constant for andrographolide‐DRP1 was 1.3 × 10⁻⁵ M (Figure 6d). As described in Figure S4A, andrographolide formed hydrogen bonds to Gln34, Ser35, Ser40, Asn246, Arg247, and Gln249 of DRP1 at binding pocket. All of these results demonstrated that andrographolide directly targeted to DRP1.

Figure 6.

Andrographolide directly bound to DRP1 and inhibited its GTPase to oppose rotenone‐induced damage. (a) Anti‐DRP1 antibody was used for immunoblotting of proteins that pulled down by biotin‐andrographolide. DRP1 pulldown was normalized with DRP1 in input. (b) Lysates from SH‐SY5Y cells were incubated with or without andrographolide (1 μM) for 9 hr, and different concentrations of pronase E was added for 20 min, DRP1 content was detected using western blot analysis. Lysates from SH‐SY5Y cells were incubated with andrographolide (0.03, 0.1, 0.3, 1 μM) for 9 hr, and a final concentration of 0.01% pronase E was added for 20 min. The DRP1 expression was detected by western blot analysis. DRP1 was normalized with actin. (c) SH‐SY5Y cells incubated with or without andrographolide (1 μM) for 9 hr, or various doses of andrographolide for 9 hr were subjected to CETSA assay. DRP1 was normalized with actin. (d) Interaction between andrographolide and DRP1 was determined via SPR analysis. (e) The DRP1‐GTPase was incubated with various doses of andrographolide, and the dissociated Pi was detected using GTPase Activity Assay Kit. (f) After 9‐hr incubation of andrographolide with rotenone‐treated SH‐SY5Y cells, the GTPase activity of DRP1 was assessed through GTP‐agarose for western blot analysis. (g) SH‐SY5Y cells transfected with si‐NC or si‐DRP1 were treated with 1‐μM andrographolide for 3 hr, followed by 30‐μM rotenone for 6 hr, and then cell viability was determined using MTT. (h, i) As a DRP‐1 inhibitor, 20‐μM Mdivi‐1 was added to SH‐SY5Y cells for 1 hr, then 1‐μM andrographolide was added for 3 hr followed by 30‐μM rotenone for 6 h, cell viability was determined using MTT, and ROS indicated by DCFH‐DA was measured. Data are mean ± SEM, n = 5. #P < .05, significantly different from control group; *P < .05, significantly different from rotenone group in a, f, g, h, and i. ΔP < .05, significantly different from Mdivi‐1 alone group. *P < .05, significantly different from control group in b, c, and e

Next, we examined whether andrographolide binding to DRP1 affected its GTPase activity. Therefore, we purified recombined DRP1‐GTPase from BL21 (DE3) (Figures S4B and 4c). As shown in Figure 6e, recombined DRP1‐GTPase in an active form hydrolyzed GTP to GDP and Pi. Andrographolide‐triggered inhibition of GTPase activity was demonstrated by decreased Pi generation. Additionally, DRP1 was pulled down by GTP‐agarose affinity chromatography from cells treated with rotenone or andrographolide. Rotenone enhanced GTPase activity of DRP1 in SH‐SY5Y cells, causing it to bind with more GTP, while andrographolide treatment down‐regulated GTPase activity of DRP1 (Figure 6f).

To further ascertain that the neuroprotective effect of andrographolide is mediated by DRP1, DRP1 was inhibited by siRNA or Mdivi‐1 in SH‐SY5Y cells. The down‐regulation of DRP1 preserved cell viability, which was the same as the effect of andrographolide. Co‐culture with andrographolide had no synergistic effect on cell viability with DRP1 knock‐down (Figure 6g), which was also ascertained using the DRP1 inhibitor Mdivi‐1 (Figure 6h). Moreover, co‐treatment with andrographolide had no synergistic effect on ROS levels with Mdivi‐1 in rotenone‐stimulated conditions, indicating andrographolide protected against rotenone through inhibiting DRP1 (Figure 6i).

Taken together, our results showed that andrographolide directly bound to DRP1, which maintained mitochondrial dynamics and exerted neuroprotective effects in PD.

4. DISCUSSION

Although levodopa remains the mainstay of treatment for PD, the side effects of levodopa, such as dyskinesia (Cenci, 2014) and oxidative stress (Langston, Ballard, Tetrud, & Irwin, 1983), point toward an urgent need for new therapeutic agents. It is known that aberrant mitochondria participate in PD, resulting in increased free radical species production, decreased mitochondrial membrane potential, reduced ATP production, disturbed calcium homeostasis, and leaked cyto c from intermembrane space (Schapira, 2008). As a quality control axis, mitochondrial biogenesis, mitochondrial dynamics (fusion–fission), and mitophagy have been suggested to be therapeutic targets in PD (Suliman & Piantadosi, 2016). Although many compounds, such as the antioxidant CoQ10, have been reported to directly preserve mitochondrial functions, no good candidate for mitochondrial medicine has achieved success to date in clinical experiments for PD therapy (Wang, Karamanlidis, & Tian, 2016). It is meaningful to discover mitochondrial protection from approved medicines. As an approved clinical effective medicine (Xi Yan Ping, Z20026249 approved by the state, Jiangxi Qingfeng Pharmaceutical Co., Ltd), andrographolide has been used in treating bronchitis, tonsillitis, and bacterial dysentery. Our previous study had shown that andrographolide can trigger mitophagy in macrophages and serve for mitochondrial quality control (Guo et al., 2014). In addition, andrographolide sulfonate can improve mitochondrial dysfunctions in the APP/PS1 mouse model of Alzheimer's disease (Geng et al., 2018). Considering the pathological role of mitochondria in PD, our study was designed to investigate the role of andrographolide on neuroprotection in MPTP‐PD, a typically mitochondria impairment associated PD model.

Along with the improved behavioural impairment, increased survival of dopaminergic neurons and decreased oxidative stress in andrographolide‐treated PD mice, the increased cell viability, decreased ROS, enhanced mitochondrial membrane potential and OXPHOS, and mitochondria‐mediated apoptosis reduction were also observed in andrographolide‐treated PD models in vivo and in vitro. Moreover, andrographolide also decreased fragmented mitochondria characterized by round, small, or markedly reduced length observed in MPTP‐ or rotenone‐treated PD models. However, mitochondrial functions and morphology complement each other, and it is still difficult to figure out which is more important. Peng et al. (2017) pointed out that excessively fragmented mitochondria lead to mitochondrial dysfunction in rotenone‐induced neurotoxicity. Further, we revealed the novel and specific function of andrographolide to protect mitochondria. The process and concept frame of this study are presented, schematically, in Figure 7 to show the novel neuroprotective mechanism of andrographolide in impaired mitochondria‐associated PD models.

Figure 7.

Concept frame and scheme of andrographolide‐mediated neuroprotective effect in Parkinson's disease. (a) Process and concept frame of this study. (b) Under normal conditions, DRP1 is located predominantly in the cytosol. Application of mitochondrial toxins such as rotenone and MPTP to simulate Parkinson's disease accelerates mitochondrial translocation of DRP1. DRP1 binds to its mitochondrial adaptors such as FIS1 and MFF and wraps around the mitochondrial tubules to drive abnormal mitochondrial division, aggravating mitochondrial malfunctions such as oxidative stress, collapsed mitochondrial membrane potential (MMP) and the depletion of ATP, and leading to apoptosis. Andrographolide can bind DRP1 directly and inhibit its GTPase activity, block the interaction between DRP1 and FIS1, and DRP1 and MFF on the mitochondria, and then inhibit excessive mitochondrial fission and cell death, resulting in the improvement of neuronal survival in Parkinson's disease

Mitochondria are highly dynamic organelles undergoing changeable morphology, which is mediated by mitochondrial fusion and fission proteins. Mitochondrial fusion mediated by MFN1, MFN2, and OPA1 allows mitochondria to exchange contents, thereby preventing permanent loss of essential components. Similar to fusion, mitochondrial fission mediated by DRP1 preserves the balance between joining and separating to maintain the mitochondrial function. Therefore, these proteins directly related to mitochondrial fusion and fission were detected in isolated mitochondria. The results showed that OPA1 and MFN2 were unchanged, while mitochondrial DRP1 was decreased by andrographolide pretreatment. These results revealed that andrographolide preserved mitochondrial morphology through inhibiting DRP1 translocation within a short time. The phosphorylation of DRP1 at Ser 616 can promote mitochondrial fission, while fission is inhibited when DRP1 is phosphorylated at Ser 637. Studies also have shown that https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1495, https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1554, activated https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1540 and https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1961/cyclin B promote phosphorylation of DRP1 at Ser 616 and increase DRP1 translocation to the surface of mitochondria (Chang & Blackstone, 2010; Li et al., 2019). Then DRP1 interacts with adaptor proteins including MiD49, MiD51, and MFF. DRP1 forms an oligomer to wrap around and constrict mitochondria into smaller and more compact organelles (Kalia et al., 2018). Bax can translocate to the mitochondrial outer membrane to form a focus and can later interact with DRP1 and MFN2 to regulate mitochondrial morphology and apoptosis, indicating that DRP1 also participates in apoptosis (Karbowski et al., 2002). FIS1 promotes DRP1 recruited from the cytoplasm to mitochondria and later interacts with DRP1 to promote mitochondrial fragmentation (Yoon, Krueger, Oswald, & McNiven, 2003). Bcl‐xl binding to DRP1 induces fission in the presence of BH3 peptides, which is independent of the mitochondrial permeability transition pore formed by Bax/Bak (Shroff et al., 2009). Dynasore is a non‐specific inhibitor of DRP1 and Mdivi‐1 is the specific inhibitor. Dynasore can inhibit the division of mitochondria via blocking GTP hydrolysis (Macia et al., 2006). Mdivi‐1‐blocked fission also blocks mitochondrial outer membrane permeabilization and inhibits Bax/Bak‐dependent apoptosis (Cassidy‐Stone et al., 2008). To confirm whether andrographolide inhibited the function of DRP1, the interactions between DRP1 and its adaptors were assessed. Biotin‐andrographolide was synthesized to pull down its potential target protein. The physical interaction between andrographolide and DRP1 was confirmed using CETSA, DARTS, and SPR assays. DRP1 harbours four distinct motifs: an N‐terminal GTPase domain binding GTP to hydrolysis to provide self‐assembling energy (Zhang & Hinshaw, 2001), a dynamin‐like middle domain for intra‐molecular interactions (Zhang, Gao, & Garavito, 2011), a short insert B domain, and a C‐terminal GTPase effector domain (GED) for membrane constriction (Pitts, McNiven, & Yoon, 2004; Zhu et al., 2004). With the physical interaction and inhibition of GTPase, the domain of DRP1‐GTPase was purified to further confirm the binding region of andrographolide. In addition, the activity of DRP1 can also be regulated by post‐translational modifications, including nitrosylation, SUMOylation, and ubiquitination (Haun, Nakamura, & Lipton, 2013). However, whether andrographolide can affect post‐translational modifications of DRP1 or bind to its other domains to regulate DRP1‐mediated fission warrants further investigation.

Although multiple mechanisms participated in activating DRP1 in PD, inhibiting fission through small molecules to rescue mitochondrial functions will be a promising therapeutic strategy. DRP1‐dominant negative mutant, DRP1‐K38A, protects dopaminergic neurons and restores dopamine release in MPTP‐PD model (Rappold et al., 2014). Additionally, a viral peptide that enhances the filaments of mitochondria can alleviate neuronal degeneration in the MPTP‐induced PD model (Szelechowski et al., 2014). The inhibition of DRP1 hyperactivation by DRP1 peptide inhibitor P110 inhibits p53‐mediated apoptotic pathways in neurons in MPTP‐PD model (Filichia, Hoffer, Qi, & Luo, 2016). Melatonin also protects cortical neurons in MPP⁺‐induced PD model by inhibiting oxidative stress and DRP1‐mediated mitochondrial fragmentation (Chuang et al., 2016). Moreover, Mdivi‐1 ameliorates VPS35 mutation and A53T‐α‐synuclein‐induced neurodegeneration (Bido, Soria, Fan, Bezard, & Tieu, 2017; Wang, Wang, et al., 2016). To confirm whether reduced DRP1‐mediated fission is involved in the neuroprotection of andrographolide, DRP1 was inhibited by siRNA and Mdivi‐1 prior to andrographolide treatment, and cell viability and ROS levels were not further significantly rescued by the combined action of DRP1 inhibition and andrographolide.

Excessive mitochondrial fission is followed by mitophagy to eliminate fragmented or damaged mitochondria (Detmer & Chan, 2007; Purnell & Fox, 2013; Zhao et al., 2017), but the relationship between DRP1 and mitophagy is complicated and DRP1‐dependent and DRP1‐independent mitophagy are observed. Prior to the encapsulation of mitochondrial fragments by expanded isolation membrane, mitophagy depends on DRP1‐mediated scission or budding and division of a mitochondrial fragment (Yamashita et al., 2016). Many researches have shown that mitophagy induced by pharmacological regulators such as urolithin A and PMI is a promising therapeutic strategy in preserving mitochondrial homeostasis and functions (East et al., 2014; Georgakopoulos et al., 2017; Ryu et al., 2016). Andrographolide slightly inhibited rotenone‐stimulated autophagy as indicated by LC3 II, but autophagosome was not observed using TEM, which is paradoxical to our previous research that andrographolide can also trigger mitophagy in ATP‐damaged macrophages, and we speculated that different cell types or environmental stressors may lead to inconsistent mitophagy.

In summary, our study demonstrated that andrographolide mitigates bradykinesia and enhances neuronal survival and mitochondrial functions, which is associated with the inhibition of mitochondrial excessive division in PD model. Andrographolide directly binds to DRP1 to inhibit the GTPase activity of DRP1, leading to the block of DRP1 oligomerization and the inhibition of excessive mitochondrial fission. These findings provide novel mechanistic insight, supporting the potential use of a clinical drug for PD treatment.

AUTHOR CONTRIBUTIONS

J.G. and W.L. performed the experiments. J.G., W.L., T.F., and W.G. analysed the data and prepared the figures. J.G., C.J., Y.S. and Z.‐H.Q. provided expertise and the mouse lines. W.G., Q.X., and J.G. conceived, directed the study, and wrote the manuscript. All authors discussed the results and commented on the manuscript.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for https://bpspubs.onlinelibrary.wiley.com/doi/abs/10.1111/bph.14207, https://bpspubs.onlinelibrary.wiley.com/doi/full/10.1111/bph.14208, and https://bpspubs.onlinelibrary.wiley.com/doi/full/10.1111/bph.14206, and as recommended by funding agencies, publishers and other organisations engaged with supporting research.

Supporting information

Figure S1 Andrographolide attenuated psychiatric comorbidities, reduced neuronal loss, and preserved mitochondrial number in MPTP‐challenged mice.(A) The mental changes were measured by open field experiments. The total distance and the time spent in center field were determined. (B) Time spent swimming was recorded during 4 minutes to analyze the depression‐like behavior in a forced swimming test. (C and D) Apoptosis and neuronal loss in striatum were measured by TUNEL staining and Nissl staining, respectively. The average number of TUNEL‐positive cells counted in five precincts. Scale bar 50 μm. Data represent mean ± SEM, n = 10 in A and B, n = 5in C.(E)The average mitochondrial numbering per picture in Figure 1 H was counted. Data represent mean ± SEM of3 mice per group, and 3 fields of striatum in each mouse were randomly photographed and calculated. #P < 0.05vs.control group; *P < 0.05vs.MPTP group.

Figure S2 Andrographolide reduced rotenone‐induced oxidative stress sand apoptosis. (A) The cytotoxiceffect of andrographolide was detected using MTT assay. (B, C) The anti‐oxidative stress effects of andrographolide on rotenone‐treated SH‐SY5Y cells were determined by MDA contents and SOD activity. (D) SH‐SY5Y cells were pretreated with andrographolide (0.1, 0.3, 1 μM) for 3 h followed by 30 μM rotenone for 6 h, bcl2, bax, caspase‐3 and cleaved caspase‐3 were examined using western blot analysis. Proteins were normalized to actin. (E) Apoptosis indicated byAnnexin V/PI staining were analyzed using flow cytometry. Data represent mean ± SEM;n = 5. #P < 0.05vs.control group; *P < 0.05vs.rotenonegroup.

Figure S3 Andrographolide promoted mitochondrial biogenesis and inhibited rotenone‐induced autophagy.(A) Transcription factors involved in mitochondrial biogenesis were analyzed using real time‐qPCR.(B) Autophagy assocaited proteinLC3wasdetermined using immunoblotting. Data are mean ± SEM;n = 5. #P < 0.05vs.control group; *P < 0.05vs.rotenonegroup.

Figure S4 The docking betweenDRP1and andrographolide, and the expression of DRP1‐GTPase under IPTG condition.(A) The potential binding sites of andrographolideto DRP1 was measured using Schrödinger Suites.(B) Schematic of the DRP1‐GTPaseconstruct. (C) The expression of DRP1‐GTPase under different IPTG induction conditions weremeasured using coomassie blue staining.

Geng J, Liu W, Gao J, et al. Andrographolide alleviates Parkinsonism in MPTP‐PD mice via targeting mitochondrial fission mediated by dynamin‐related protein 1. Br J Pharmacol. 2019;176:4574–4591. 10.1111/bph.14823