Intranasal insulin improves mitochondrial function and attenuates motor deficits in a rat 6‐OHDA model of Parkinson's disease

Abstract Aims Experimental and clinical evidences demonstrate that common dysregulated pathways are involved in Parkinson’s disease (PD) and type 2 diabetes. Recently, insulin treatment through intranasal (IN) approach has gained attention in PD, although the underlying mechanism of its potential therapeutic effects is still unclear. In this study, we investigated the effects of insulin treatment in a rat model of PD with emphasis on mitochondrial function indices in striatum. Methods Rats were treated with a daily low dose (4IU/day) of IN insulin, starting 72 h after 6‐OHDA‐induced lesion and continued for 14 days. Motor performance, dopaminergic cell survival, mitochondrial dehydrogenases activity, mitochondrial swelling, mitochondria permeability transition pore (mPTP), mitochondrial membrane potential (Δψm), reactive oxygen species (ROS) formation, and glutathione (GSH) content in mitochondria, mitochondrial adenosine triphosphate (ATP), and the gene expression of PGC‐1α, TFAM, Drp‐1, GFAP, and Iba‐1 were assessed. Results Intranasal insulin significantly reduces 6‐OHDA‐induced motor dysfunction and dopaminergic cell death. In parallel, it improves mitochondrial function indices and modulates mitochondria biogenesis and fission as well as activation of astrocytes and microglia. Conclusion Considering the prominent role of mitochondrial dysfunction in PD pathology, IN insulin as a disease‐modifying therapy for PD should be considered for extensive research.


| INTRODUC TI ON
Parkinson's disease (PD) is the second common neurodegenerative disorder associated with progressive loss of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNpc) and degeneration of projecting dopaminergic terminals to the striatum, which is linked to motor deficits. 1 PD has a multifactorial pathogenesis, involving both genetic and environmental factors. 1,2 The most important mechanisms implicated in development of PD include aggregation of misfolded proteins, mitochondrial dysfunction, energy failure, oxidative stress, and dysregulated calcium homeostasis. 3 Current research extensively focuses on finding new early diagnostic approaches and introducing molecular targets for disease-modifying treatments specified for underlying causes of PD and reducing DA neuronal loss. [4][5][6] Regarding the neuroprotective effects of insulin in the brain, 7 and the probable causative role of insulin resistance in PD pathogenesis, 8 emerging studies continue to demonstrate the beneficial effects of intranasal (IN) insulin in both animal models of PD and humans in clinical trials. [9][10][11] The insulin receptors are widely found in basal ganglia and SN and insulin is described as a critical regulator of energy balance, neuronal survival, and growth, neurotransmission, and maintenance of synapses in brain. 12 However, little is known about the involving mechanisms in neuroprotective effects of insulin in PD.
Nigrostriatal neurons have a high energy requirement and dysfunction of mitochondria, as the source of energy supply, plays a key role in the development of PD. 13,14 Mitochondrial dysfunction manifested by reduced adenosine triphosphate (ATP) production and calcium buffering capacity, as well as impaired degradation of damaged mitochondria through mitophagy and replacement with new functional mitochondria through biogenesis, has been linked with PD. 15 Mitochondrial Ca +2 overloading activates the opening of mitochondrial membrane transition pore (mPTP) and consequently causes the collapse of mitochondrial membrane potential (Δψ m ), reduced ATP level, release of pro-apoptotic mediators into the cytosol and finally mitochondria-mediated cell death. 14 Furthermore, mitochondrial dynamic (fission and fusion) and biogenesis have essential roles in neuronal survival. 14,16 The main factor involved in mitochondrial fission is dynamin-related protein 1 (Drp-1), a dynamin-like GTPase, which is recruited to the fission sites of outer membrane of mitochondria. 16 Production of new mitochondria is principally regulated by peroxisome proliferator-activated receptor gamma (PPARγ) coactivator 1α (PGC-1α) which activates the expression of transcription factors like nuclear respiratory factors (NRF1 and 2) and mitochondrial genes encoded by nucleus such as mitochondrial transcription factor A (TFAM). 17 Emerging evidence suggests mitochondria as a novel and highly relevant therapeutic target to reduce neurodegeneration in PD. 18 Inadequate insulin signaling or impaired cellular response to insulin, termed as insulin resistance, is among the causes of mitochondrial dysfunction in the development and progression of PD. 19 Insulin act as a modulator of electron transport chain activity and the main regulator of mitochondrial biogenesis, through activation of the PI3K/Akt pathway. 20 In this study, we investigated the effects of IN insulin administration on various indices of mitochondrial function in the striatum of 6-OHDA PD modeled rats. In parallel, mitochondrial biogenesis and mitochondrial fission, behavioral motor performance, DA neurons survival, and glial cells activation markers were also assessed.

| Animals
Male Wistar rats (300-325 g) from the breeding colony of

| Stereotaxic surgery
One hour before the surgery, desipramine-HCl (25 mg/kg) was injected intraperitoneal (i.p.) to prevent uptake of 6-OHDA into noradrenergic terminals. 21 Then the rats were anesthetized with ketamine/ xylazine (80/20 mg/kg, i.p.) and were placed on a stereotaxic frame with a rat adaptor. A total of 20 μg 6-OHDA (in 4 μl normal saline with 0.2 mg/ml ascorbic acid) was injected into the right medial forebrain The control rats received the same volume of vehicle (normal saline with 0.2 mg/ml ascorbic acid).

| Insulin administration and experimental groups
From day three after surgery, rats received regular human insulin (4IU/day; 2IU in each nostril, for 14 days) or normal saline intranasally in an upright head position, as previously described. 23

| Behavioral tests
All rats were trained on the behavioral tests (narrow beam and rotarod tests) for two consecutive days before stereotaxic surgery. One day after the last insulin/normal saline administration, the rats were subjected to beam, rotarod and apomorphine-induced rotational tests.

| Apomorphine-induced rotation
The rats were placed in the chamber to habituation. After 30 min, they

| Rotarod test
Motor performance was evaluated using a rotarod apparatus as previously described. 24 The rats were trained five times a day for two consecutive days, before stereotaxic surgery, until they could stay on the rotating rod for 300 s. In the first training day, the rotarod speed was constant at 10 rpm, and in the second day, it was accelerating from 5 to 20 rpm during 300 s. In the test sessions, the speed was increased from 5 to 40 rpm over 300 s, and each rat performed five trials with 300 s cutoff. The average of five trials was considered as the final score.

| Narrow beam test
Narrow beam apparatus is a long wooden beam (100 cm in length, 4 cm wide and 3 cm thick) elevated 80 cm above the ground. A line is drawn 20 cm from the start end of beam, and a cage is placed at the other end. During the training and testing sessions, the rats were placed entirely within the 20 cm starting zone facing its home cage, and the total time of walking on the beam to reach the home cage was recorded. Before the surgery, the animals received two consecutive days of training each consisting of five trials. Each rat was subjected to five trials with 120 s cutoff in testing sessions, and the average of five trials was considered as the final score. 25

| Immunohistochemistry
The rats were anaesthetized with ketamine/xylazine (80/20 mg/ kg) and transcardially perfused with normal saline, followed by 4% PFA. The brains' right hemisphere (ipsilateral to lesion) were removed, and kept in the same fixative for 24 h, and in 30% sucrose for 72 h. Afterward, each brain is placed on to a pre-labeled tissue base mold, embedded with optimal cutting temperature compound (OCT) and stored at −80°C. Coronal sections (10 μm) were cut on a cryostat and used for immunostaining. In brief, frozen sections were left at room temperature for 10-20 min. Then they were put in methanol for 30 s and washed in distilled water for few seconds. Endogenous peroxidase was inactivated by 3% aqueous solution of hydrogen peroxide, and 10% normal goat serum (30 min) was used to block non-specific binding sites. The sections were incubated with rabbit polyclonal anti-TH antibody (1:1000) over night at 4°C. TH-immunoreactivity was detected by HRP conjugated secondary antibody and visualized using liquid DAB followed by counterstaining with Mayer's hematoxylin.
The boundary between SNpc and ventral tegmental area (VTA) was determined at ×40 magnification, and the immunoreactive neurons were counted at ×400 magnification. Three animals were used in each group, and the values from at least three sections were averaged for each animal.

| Extraction of striatal mitochondria
The mitochondria were isolated from right striatum (ipsilateral to lesion) as previously described. 26,27 Briefly, the animals were anesthetized using ketamine/xylazine (80/20 mg/kg, i.p.) and the striatum was rapidly dissected from the brain on ice and washed with cold normal saline. The tissues were homogenized in a buffer containing 70 mM mannitol, 220 mM sucrose, 0.5 mM EGTA, 2 mM HEPES, 0.1% essentially fatty acid-free BSA (pH = 7.4) at a 10:1 ratio of isolation buffer to striatum tissue (v:w). Then, the homogenate was centrifuged at 600g (10 min at 4°C) to remove cells debris. The supernatants were additionally centrifuged at 10,000g (10 min at 4°C) to precipitate the mitochondria. This step was repeated at least three times by fresh isolation buffer. The obtained mitochondrial fraction was resuspended into the same buffer as above but without BSA, at a final concentration of 20mg protein/ml, determined by the Bradford reagent.

| Mitochondrial swelling and permeabilization
Mitochondrial swelling was evaluated by the light scattering method as described before. 26

| Mitochondrial membrane potential
The uptake of the cationic fluorescent probe, Rhodamine 123, was used to estimate of mitochondrial depolarization. 26

| Reactive oxygen species (ROS) formation
The fluorescent probe DCFH-DA was used to evaluate ROS formation in mitochondria. 26

| Mitochondrial GSH content
Mitochondrial GSH level was determined using DTNB, as an indicator of GSH, by a spectrophotometric method. 27 The mitochondrial suspension (0.5 mg protein/ml) was treated with TCA to a final concentration of 10% (w/v), and centrifuged at 15,000g (1 min at 4ºC) to remove denatured proteins. Afterward, 100 μl of DTNB (0.04% in phosphate buffer) was added, and intensity of produced yellow color was measured at 412 nm using the multifunctional microplate reader.

| Mitochondrial ATP content
The ATP level of mitochondria was measured using a luciferaseluciferin based kit (ENLITEN ® from Promega, Madison, WI). 28 The samples and buffer solutions were prepared according to the kit instructions. Briefly, isolated mitochondria (1 mg protein/ml) were treated with 100 μl TCA (0.5% w: v) and centrifuged (10,000g, 10 min, 4°C). Then, 100 μl of the supernatant was mixed with 100 μl of ATP kit content. The luminescence intensity was detected at 560 nm using the multifunctional microplate reader.

| RNA isolation and qPCR Protocol
The rats were decapitated and right striatum (ipsilateral to lesion) was immediately dissected on ice, and total RNA was extracted

| Statistical analysis
Statistical analyses were performed using 6th version of GraghPad Prism. Data are expressed as mean ± standard error of mean (SEM).
Kolmogorov-Smirnov (KS) normality test indicated that the data have normal distribution; therefore, comparisons were done using parametric tests, two-way analysis of variance (ANOVA) with repeated measures and one-way ANOVA followed by Tukey's post hoc. Statistical significance was set at p < 0.05.

| RE SULT
3.1 | Intranasal insulin administration had no effect on the body weight and blood glucose levels

| The effect of intranasal insulin on behavioral performance
Statistical analysis revealed that there was a significant difference reduce it compared to 6-OHDA (p < 0.001), however, no difference was observed between 6-OHDA and saline-treated groups (p > 0.05) ( Figure 2C).

| The effects of insulin on the expression of genes involved in mitochondrial biogenesis and fission in striatum
However, it was significantly higher in 6-OHDA+Ins group in comparison to 6-OHDA (p < 0.01) ( Figure 5A). ANOVA analysis indicated that 6-OHDA increased TFAM gene expression compared to sham (p < 0.001), and insulin could significantly attenuate it (p < 0.001) in comparison to 6-OHDA ( Figure 5B). Tukey's post hoc test indicated that although 6-OHDA decreased Drp-1 gene expression compared to sham (p < 0.05) insulin significantly enhanced it compared to both sham and 6-OHDA groups (p < 0.05 and p < 0.001, respectively) ( Figure 5C).

| DISCUSS ION
In current study, we found that IN insulin could rescue dopaminergic neurons against cell death induced by 6-OHDA, and ameliorate motor deficits in a rat model of PD, which is consistent with previous studies. 9,11 6-OHDA undergoes oxidation inside the cells, produces reactive oxygen species (ROS), and results in mitochondrial enzymes dysfunction, mtDNA mutation, disruption of mitochondrial membrane permeability and apoptotic cell death. 29,30 Also, excessive free radicals induced by 6-OHDA may open the mPTP, 31 and drive some antioxidant molecules such as GSH to exit from mitochondria. Therefore, ability of the mitochondria will not be enough to neutralize ROS, which in turn results in more ROS production.
Furthermore, induction of mPTP increases mitochondrial membrane permeability and causes mitochondria to become more depolarized, therefore, Δψ m will be abolished resulting outflow of protons across the outer mitochondrial membrane and eventually disruption in ATP generation. 14 Similarly, we observed increased ROS formation and mitochondrial membrane permeability, decreased GSH content and ATP level in striatum of 6-OHDA-lesioned rats. Interestingly, we observed that treatment with insulin attenuated all detrimental effects of 6-OHDA. Previous studies have also found that insulin decreases ROS production in the neurons 32,33 and it has an important role in enhancement of Δψ m , intracellular ATP levels and NADPH redox state, through activation of PI3K/Akt/CREB pathway, and therefore, it can improve axonal outgrowth. 34 Producing the new mitochondria is controlled by mitochondrial biogenesis pathway. 35 In the context of neural damage, mitochondrial biogenesis can be activated as a compensatory mechanism to protect the neurons. It has been reported that excessive production of ROS increases PGC-1α in neurons after ischemia, which It has been indicated that astrocytes may release extracellular mitochondrial particles that could enter into neurons to improve the viability of the cells, and therefore, increase the recovery after stroke. 47 In addition, astrocytes are essential components of antioxidant defense system in the brain which regulate extracellular concentrations of glutamate and antioxidant compounds. 48 Therefore, increase in astrocyte function may also contribute to decrease oxidative damage in PD which has been previously reported. 49 Astrocytes are also an integral component of the antioxidant defense system in the brain through the regulation of extracellular glutamate concentrations and production of antioxidant compounds. 48 Microglial activation was also observed after insulin treatment, characterized by increased Iba-1 gene expression in striatum. It has F I G U R E 6 The expression of genes involved in astrocyte and microglia activity in striatum. Although 6-OHDA had no effect on (A) GFAP and (B) Iba-1 mRNA levels, as astrocyte and microglia activity markers, insulin significantly increased them. Data are expressed as means ± SEM (n = 3/group). *** p < 0.001 vs. Sham. ### p < 0.001 vs. 6-OHDA been proposed two subpopulation of active microglia, M1 phenotype and M2 phenotype. 58 M1 microglia activation leads to production of several pro-inflammatory cytokines such as ROS and tumor necrosis factor-a (TNF-α). Conversely, M2 microglia activation alleviates inflammatory response and promotes tissue repair by releasing the anti-inflammatory cytokines include interleukin-4 (IL-4), IL-10 and various trophic factors such as TGF-β. 59 Recently, some increasing evidence has revealed that PI3K/Akt pathway plays an important role in microglial M1/M2 polarization via induction of M2-type cell accumulation and the inhibition of M1-type microglia production. 60,61 A study also showed that insulin treatment reduces the release of pro-inflammatory cytokine TNF-α 62 and induces an increase in the TGF-β receptors at the cell surface which causes enhancement in the cell responsiveness to autocrine TGF-β. 63 Although the molecular mechanisms of TGF-β to promote anti-inflammatory effects of microglia are less understood, an in vitro study has been shown that TGF-β1 (an isoform of TGF-β) might regulate different states of microglia activation by inhibition of M1 form, and shifting toward M2 phenotypes. 64 Therefore, insulin likely upregulates anti-inflammatory phenotypes of reactive glia via PI3K/Akt pathway which it may be at least a part of neuroprotective effects of insulin.

| CON CLUS ION
The findings of the present study demonstrated that IN administra-

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

AUTH O R CO NTR I B UTI O N
Conception and design of the study: Abolhassan Ahmadiani,

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable requests.