Effect of Mitoquinone (Mito-Q) on Neuropathic Endpoints in an Obese and Type 2 Diabetic Rat Model

Abstract

This study sought to determine whether the addition of mitoquinone (Mito-Q) in the diet is an effective treatment for peripheral neuropathy in animal models of diet-induced obesity (pre-diabetes) and type 2 diabetes. Unlike other anti-oxidative stress compounds investigated as a treatment for peripheral neuropathy, Mito-Q specifically targets mitochondria. Although mito-Q has been shown to reduce oxidative stress generated by mitochondria there have been no studies performed of the effect of Mito-Q on peripheral neuropathy induced by diet-induced obesity or type 2 diabetes. Diet-induced obese (12 weeks after high fat diet) or type 2 diabetic rats (12 weeks of high fat diet and 4 weeks after the onset of hyperglycemia) were treated via the diet with Mito-Q (0.93g/kg diet) for 12 weeks. Afterwards, glucose utilization, vascular reactivity of epineurial arterioles to acetylcholine and peripheral neuropathy related endpoints were examined. The addition of Mito-Q to the diets of obese and diabetic rats improved motor and/or sensory nerve conduction velocity, cornea and intraepidermal nerve fiber density, cornea sensitivity and thermal nociception. Surprisingly, treating obese and diabetic rats with Mito-Q did not improve glucose utilization or vascular reactivity by epineurial arterioles to acetylcholine. These studies imply that mitochondrial dysfunction contributes to peripheral neuropathy in animal models of pre-diabetes and late-stage type 2 diabetes. However, improvement in peripheral neuropathy following treatment with Mito-Q was not associated with improvement in glucose utilization or vascular reactivity of epineurial arterioles to acetylcholine.

Introduction

Our work and others have demonstrated that two of the primary mechanisms responsible for peripheral neuropathy in obesity and type 2 diabetes are increased oxidative and inflammatory stress [1–8]. However, numerous pathways can contribute to increasing oxidative and inflammatory stress in obesity and type 2 diabetes and knowing which of these to target in order to prevent or slow progression of peripheral neuropathy is a challenge. Abnormal mitochondrial activity and dysfunction has been linked to a number of metabolic diseases including obesity and type 2 diabetes [3,9]. Mitochondrial dysfunction and increased oxidative stress have also been linked to insulin resistance and peripheral neuropathy [6,8,10]. Many studies have examined the effect of anti-oxidants on diabetic peripheral neuropathy but the agents used did not specifically target the mitochondria. We have previously shown that tempol and α-lipoic acid can improve peripheral neuropathy in high fat fed and type 2 diabetic rodents [11–14]. We have also shown that M40403, a manganese(II) complex with a bis(cyclo-hexylpyridine)-substituted macrocyclic ligand that was designed to be a selective functional mimetic of superoxide dismutase, improved vascular and neural endpoints and reduced markers of oxidative stress in the vasculature in a type 1 diabetic rat model [15]. Here we report on studies that examined the effect of mitoquinone (Mito-Q) on peripheral neuropathy caused by diet-induced obesity and type 2 diabetes. MitoQ is orally active, can cross mammalian membranes, has mitochondrial uncoupling properties and can increase energy expenditure [16]. Mito-Q has been shown to reduce oxidative stress generated by the mitochondria and is particularly effective against lipid peroxidation [17]. It has also been shown to prevent weight gain in high fat fed mice, ameliorate hepatic dysfunction, attenuate liver fibrosis and improve glucose utilization in obese mice and aortic stiffening in old mice [1,18,19]. Mito-Q may also have neuroprotective effects and be advantageous against neuroinflammatory diseases such as multiple sclerosis and amyotrophic lateral sclerosis [20]. However, there have been no studies of the effect of Mito-Q on peripheral neuropathy induced by diet-induced obesity or type 2 diabetes. The goal of these studies was to fill this void and determine whether treating animal models of diet-induced obesity, thought to be a pre-diabetic state, and type 2 diabetes with an anti-oxidant that targets mitochondria can prevent or slow the progression of peripheral neuropathy.

Materials and Methods

Reagents and Supplies:

All reagents required for the synthesis of Mito-Q were purchased from commercial sources and used without further purification. Mito-Q was synthesized following a previously described approach from the commercially available intermediate, idebenone, [21] using methods as previously described [22]. MitoQ-β-cyclodextrin complex was prepared and standardized experimentally as previously reported [22].

Animals, Diets and Experimental Design:

Male Sprague-Dawley (Harlan Sprague Dawley, Indianapolis, IN) rats 10-11 weeks of age were housed in a certified animal care facility and food (Harlan Teklad, #7001, Madison, WI) and water were provided ad libitum. All institutional (ACORP #1691101) and NIH guidelines for use of animals were followed. At 12 weeks of age rats were divided into 5 groups. One group was designated as the control group and remained on the standard diet, which contained 4.25 gm% as fat, for the entire experimental period. The other 4 groups were placed on a high fat diet (D12451; Research Diets, New Brunswick, NJ). The high fat diet contained 24 gm% fat, 24 gm% protein and 41 gm% carbohydrate. The primary source of the increased fat content in the diet was lard. After 8 weeks, two groups were treated with a low dose of streptozotocin (30 mg/kg) to induce hyperglycemia [23]. All four groups remained on the high fat diet for an additional 4 weeks. Subsequently, the high fat diet of one of the high fat fed groups and one of the diabetic groups was supplemented with 0.93 g/kg diet of Mito-Q (368.3 μmol Mito-Q/g) (D18071601; Research Diets, New Brunswick, NJ) [24]. The Mito-Q was synthesized as described above and provided to Research Diets. All groups remained on the high fat diet for a total treatment period of 12 weeks. The amount of diet consumed by rats in each of the five groups was: Control 51.7 ± 1.7 g/day/kg rat, Obese 29.8 ± 1.6 g/day/kg rat, Obese + Mito-Q 30.3 g/day/kg rat, Diabetic 57.8 ± 3.3 g/day/kg rat, and Diabetic + Mito-Q 56.8 ± 2.7 g/day/kg rat.

Neural and Vascular Endpoints:

In order to fully determine the impact of Mito-Q on neural and vascular complications associated with obesity and type 2 diabetes multiple endpoints were examined. Glucose utilization, thermal nociceptive latency of the hindpaws and cornea sensitivity was performed in un-anesthetized rats. Motor and sensory nerve conduction velocity and density of corneal nerves by corneal confocal microscopy were determined in rats anesthetized with sodium pentobarbital (50 mg/kg, i.p., Abbott Laboratories, North Chicago, IL). Following these procedures the anesthetized rats were euthanized by exsanguination and tissues were harvested to determine intraepidermal nerve fiber density of the hindpaw and vascular reactivity of epineurial arterioles of the sciatic nerve to acetylcholine. These outcome measures were determined by procedures routine in our laboratory, as we previously published in detail [14].

Physiological markers:

Non-fasting blood glucose was determined with Aviva Accu-Chek strips. Serum was collected to determine thiobarbituric acid reactive substances as previously described [7]. Lipid peroxide (LPO) assay kit by Cayman Chemicals (Ann Arbor, MI, USA) was used according to the manufactures instructions without modifications to determine lipid hydroperoxides in liver [18].

Data Analysis:

Results are presented as mean ± SEM. Comparisons between the groups were conducted using one-way ANOVA and Bonferroni posttest comparison (Prism software; GraphPad, San Diego, CA). Concentration response curves for acetylcholine were compared using a two-way repeated measures analysis of variance with autoregressive covariance structure using proc mixed program of SAS [5–7]. A p value of less than 0.05 was considered significant.

Results

Data in Table 1 demonstrate that all rats in the 5 different groups weighed about the same when they were entered into the study. When rats were weighed at the conclusion of the study, rats in the obese and obese + Mito-Q groups weighed significantly more than rats in the control group or in the diabetic and diabetic + Mito-Q groups. As expected, rats in the diabetic and diabetic + Mito-Q groups were hyperglycemic. Moreover, treatment with Mito-Q did not influence the final blood glucose level compared to untreated diabetic rats.

Table 1:

Effect of Treatment of Diet-induced Obese or Type 2 Diabetic Rats with Mito-Q on Body Weight, Blood Glucose, Serum Thiobarbituric Acid Reactive Substances and Liver Lipid Hydroperoxides

Determination	Control	Obese	Obese + Mito-Q	Diabetic	Diabetic + Mito-Q
	(9)	(8)	(9)	(9)	(9)
Start weight (g)	327 ± 2	323 ± 4	320 ± 7	329 ± 3	325 ± 3
Final weight (g)	468 ± 7	536 ± 13a	551 ± 13a	469 ± 22b	416 ± 12b
Blood glucose (mg/dl)	142 ± 7	167 ± 9	141 ± 2	384 ± 38a,b	440 ± 31a,b
Thiobarbituric acid reactive substances (μg/ml)	0.9 ± 0.09	1.04 ± 0.07	0.97 ± 0.07	1.24 ± 0.08	1.12 ± 0.13
Lipid Hydroperoxides (nmol/g liver)	21.7 ± 3.1	57.8 ± 7.8a	42.4 ± 5.8	62.5 ± 3.5a	45.4 ± 3.5a

Data are presented as the mean ± S.E.M.

^aP < 0.05 compared to control rats;

^bP < 0.05 compared to obese rats;

^cP < 0.05 compared to diabetic rats.

Parentheses indicate the number of experimental animals.

Two generic markers of oxidative stress serum thiobarbituric acid reactive substances and liver lipid hydroperoxides were measured to determine the efficacy of treatment of Mito-Q. Data in Table 1 demonstrate that serum thiobarbituric acid reactive substances are elevated in obese and to a greater extent in diabetic rat but not significantly and treatment with Mito-Q improves these modest changes this. Liver lipid hydroperoxides are also increased in obese and diabetic rats. Treatment of obese and diabetic rats with Mito-Q reduces the increase in liver lipid hydroperoxides but in both obese and diabetic rats their levels remain elevated.

Figure 1 provides data for glucose utilization. As previously reported glucose utilization by diet-induced obese rats and to greater extent type 2 diabetic rats is significantly impaired compared to control rats [25]. Treating obese rats with Mito-Q improved glucose utilization with area under the curve (AUC) for obese + Mito-Q rats not significantly different from control rats. In contrast, treating diabetic rats with Mito-Q did not improve glucose utilization.

Figure 1.

Effect of Mito-Q treatment of diet-induced obese and type 2 diabetic rats on glucose utilization. Rats were fasted overnight and injected with a solution containing 2g/kg glucose (i.p.). Circulating glucose levels were determined immediately prior to the glucose injection and then for the next 15 to 240 min. Area under the curve was determined for each of the five groups of rats studied (Control 43838 ± 3458; Obese 69875 ± 2997*; Obese + MitoQ 52503 ± 2874⁺; Diabetic 115118 ± 6317*^,+; Diabetic + MitoQ 122115 ± 2005*^,+, arbitrary units. * P < 0.05 compared to control rats; + P < 0.05 compared to obese rats. Data are presented as the mean ± S.E.M. for glucose utilization in mg/dl. The number of rats in each group was the same as presented in Table 1.

Different endpoints related to peripheral nerve function were determined (Table 2). Motor nerve conduction velocity was significantly decreased in diabetic rats compared to control rats and significantly improved when diabetic rats were treated with Mito-Q. Sensory nerve conduction velocity was significantly decreased in both obese and diabetic rats compared to control rats. Treating these rats with Mito-Q improved sensory nerve conduction velocity, which was no longer significantly different from control rats. Decrease in density of small sensory nerve fibers in the epidermis of the skin and cornea has been promoted as an early marker of peripheral neuropathy and has been reported to occur in human subjects with insulin resistance as well as diabetes [26]. In this study intraepidermal nerve fiber density and cornea nerve fiber length were significantly decreased in both diet-induced obese rats and type 2 diabetic rats. Treating obese rats with Mito-Q did not improve intraepidermal nerve fiber density but did significantly improve cornea nerve fiber length. Treating type 2 diabetic rats with Mito-Q significantly improved intraepidermal nerve fiber density and cornea nerve fiber length. However, both intraepidermal nerve fiber density and cornea nerve fiber length in Mito-Q treated diabetic rats remained significantly decreased compared to control rats.

Table 2:

Effect of T reatment of Diet-induced Obese or Type 2 Diabetic Rats with Mito-Q on Motor and Sensory Nerve Conduction Velocity, Intraepidermal and Cornea Nerve Fiber Density, Thermal Nociception and Cornea Sensitivity

Determination	Control	Obese	Obese + Mito-Q	Diabetic	Diabetic + Mito-Q
	(9)	(8)	(9)	(9)	(9)
MNCV (m/sec)	50.1 ± 1.6	52.7 ± 1.4	51.2 ± 1.7	40.9 ± 1.4a,b	48.9 ± 2.9c
SNCV (m/sec)	37.2 ± 1.4	30.9 ± 1.1a	34.6 ± 0.6	30.1 ± 0.8a	33.8 ± 0.6
IENF (profiles/mm)	21.8 ± 0.4	17.5 ± 0.4a	17.8 ± 0.2a	14.0 ± 0.2a,b	17.8 ± 0.4a,c
CNFL (mm/mm2)	8.1 ± 0.2	4.5 ± 0.4a	7.2 ± 0.4b	4.2 ± 0.2a	6.1 ± 0.4a,c
Thermal nociception (sec)	11.3 ± 0.5	17.2 ± 0.7a	12.1 ± 0.4b	20.1 ± 1.4a	13.3 ± 0.8c
Cornea sensitivity (cm)	5.6 ± 0.1	4.3 ± 0.1a	4.8 ± 0.2a	4.3 ± 0.1a	5.1 ± 0.2c

Data are presented as the mean ± S.E.M.

^aP < 0.05 compared to control rats;

^bP < 0.05 compared to obese rats;

^cP < 0.05 compared to diabetic rats.

Parentheses indicate the number of experimental animals.

We also examined responsiveness of hindpaw of rats to a thermal stimulus. Both diet-induced obese rats and type 2 diabetic rats were hypoalgesic at the end of the study period. Treating obese and diabetic rats with Mito-Q significantly improved thermal nociception.

Cornea sensitivity was determined using a Cochet-Bonnet esthesiometer. Corneal sensitivity was significantly impaired in obese and type 2 diabetic rats. Treating obese and diabetic rats with Mito-Q modestly improved cornea sensitivity although sensitivity remained significantly decreased in obese rats treated with Mito-Q compared to control rats.

We have previously reported that impairment of vascular relaxation to acetylcholine occurs early in type 1 diabetic rats preceding slowing of motor nerve conduction velocity and is also impaired in diet-induced obese rats and type 2 diabetic rats [23,27,28]. Data in Figure 2 demonstrated that vascular relaxation to acetylcholine was impaired in both obese and type 2 diabetic rats. Treating diabetic rats with Mito-Q modestly improved relaxation to acetylcholine but the difference between diabetic treated and untreated rats was not significant while Mito-Q treatment of obese rats had no effect improving vascular relaxation to acetylcholine.

Figure 2.

Effect of Mito-Q treatment of diet-induced obese and type 2 diabetic rats on vascular relaxation by acetylcholine in epineurial arterioles of the sciatic nerve. Pressurized arterioles (40 mm Hg and ranging from 60-100 μm luminal diameters) were constricted with phenylephrine (30-50%) and incremental doses of acetylcholine were added to the bathing solution while recording steady state vessel diameter. The groups examined and the number of rats in each group was the same as presented in Table 1. Data are presented as the mean of % relaxation ± S.E.M. * P < 0.05 compared to control rats; + P < 0.05 compared to Diabetic.

Discussion

The primary finding from this study was that Mito-Q improved multiple endpoints associated with peripheral neuropathy in both diet-induced obese and type 2 diabetic rats. This improvement occurred even though acetylcholine-mediated vascular relaxation by epineurial arterioles that provide circulation to the sciatic nerve was not significantly improved. This is the first pre-clinical study conducted demonstrating significant improvement in multiple neuropathic endpoints without correction of neurovascular dysfunction. This outcome highlights the complexity of the etiology of peripheral neuropathy.

Previously, we have demonstrated that intervention with different compounds with antioxidant properties improved vascular dysfunction in diabetic rat models as well as peripheral neuropathy. Treating type 1 diabetic rats with M40403, a superoxide dismutase mimetic, daily subcutaneously following a prevention protocol improved decrease in endoneurial blood flow, acetylcholine-mediated vascular relaxation by epineurial arterioles and motor nerve conduction velocity [15]. The treatment protocol also reduced the appearance of superoxide in the aorta and superoxide and peroxynitrite in epineurial arterioles. We have also examined the effect of dehydroepiandrosterone, a naturally occurring adrenal androgen with antioxidant properties. Using a prevention protocol in type 1 diabetic rats we found that dietary supplementation with dehydroepiandrosterone also protected diabetes-induced impairment of endoneurial blood flow, acetylcholine-mediated vascular relaxation by epineurial arterioles and motor nerve conduction velocity [29]. In yet another study we used a potent peroxynitrite decomposition catalyst Fe(In)tetrakis-2-(N-triethylene glycol monomethyl ether) pyridyl porphyrin (FP15) dissolved in drinking water to treat type 1 diabetic rats for 4 weeks following 2 weeks of no treatment [30]. At the end of 6 weeks untreated diabetic rats had impaired motor and sensory nerve conduction velocities, reduced endoneurial blood flow, decreased acetylcholine-mediated relaxation by epineurial arterioles and increased nitrotyrosine immunofluorescence in sciatic nerve, dorsal root ganglion and epineurial arterioles. FP15 treatment in a dose dependent manner alleviated all of these deficits.

We have also extensively studied the effect of α-lipoic acid on vascular and neural dysfunction in type 1 and type 2 diabetic rats. α-Lipoic acid is a good metal chelator and capable of scavenging hydroxyl radicals, hypochlorous acid, and singlet oxygen, but not superoxide or peroxyl radicals [31–34]. However, in its reduced form, as dihydrolipoic acid, it is a good scavenger of superoxide and prevents initiation of lipid peroxidation [31–34]. In vivo, α-lipoic acid can be converted into dihydrolipoic acid [34, 35]. This reaction requires either NADH via a mitochondrial pathway or NADPH via cytosolic pathways [34, 35]. The latter cofactor is reduced in diabetes due to the increased flux of glucose through the aldose reductase pathway [36, 37]. In our studies with type 1 diabetic and two different rat models of type 2 diabetes, Zucker diabetic fatty rat and high fat fed low dose streptozotocin treated rats, α-lipoic acid treatment reversed multiple neuropathic endpoints and improved vascular reactivity [12,13,38]. Moreover, we have shown that treating type 1 diabetic rats with a combination of α-lipoic acid and an aldose reductase inhibitor resulted in increased circulating levels of dihydrolipoic acid and improvement in neural and vascular dysfunction [39]. In a study using isolated epineurial arterioles from diabetic rats we demonstrated that pre-treating these blood vessels in vitro with dihydrolipoic acid, tempol or arginine restored acetylcholine-mediated vascular relaxation [40]. This study suggested that increased oxidative stress and reduction in nitric oxide availability is responsible for the impairment in endothelium-dependent vasodilation observed in epineurial arterioles from diabetic rats. Overall, these studies imply that an increase in oxidative stress contributes to vascular and neural complications observed in diabetic rats. However, the source of the reactive oxygen species leading to vascular and neural impairment is unknown.

Oxidative stress has been defined as a disturbance in the balance between the production of reactive oxygen species (oxygen free radicals, i.e., hydroxyl radical [OH·], superoxide anion [O₂⁻], and hydrogen peroxide [H₂O₂]) and antioxidant defenses [41]. In this study we have shown that reduction of oxidative damage generated by the mitochondria following treatment with Mito-Q improved neural dysfunction but not vascular impairment. Mito-Q differs from other anti-oxidative stress agents that we have studied in that Mito-Q mitigates oxidative damage (mainly lipid peroxidation) rather than radical production. The finding that Mito-Q did not significantly improve vascular dysfunction in obese and type 2 diabetic rats as determined by measuring vascular relaxation to acetylcholine by epineurial arterioles of the sciatic nerve suggests that the generation of reactive oxygen species responsible for vascular dysfunction in these models is more likely due to NAD(P)H oxidase, xanthine oxidases, lipoxygenase, cyclooxygenase, or uncoupled nitric oxide synthase and reduced antioxidant mechanisms rather than the mitochondria [42]. However, our studies did not address the impact treatment with Mito-Q of obese or diabetic rats may have on properties of endoneurial vessels. Endoneurial vessels are important for maintaining hemodynamic properties of peripheral nerves and abnormalities of the endoneurial microvasculature has been linked to peripheral nerve dysfunction in diabetes [43,44]. It is possible that Mito-Q may preserve the microcirculation of endoneurial vessels and/or reactivity to other vasodilators such as calcitonin gene-related peptide.

In the neuron several laboratories have demonstrated that diabetes decreases activity in the mitochondrial respiratory chain in dorsal root ganglion of rats and mice, a process that could switch maintenance of neuronal bioenergetics away from the mitochondrial electron transport chain toward less efficient anaerobic metabolic pathways [6]. These reports contradict the more conventional point of view that hyperglycemia leads to increased mitochondrial oxidative phosphorylation leading to increased production of reactive oxygen species that exceeds the neutralization capability of detoxification pathways such as superoxide dismutase, catalase and glutathione [45]. One well recognized signal of oxidative/nitrosative stress in neuron and Schwann cells is activation of the poly(ADP-ribose) polymerase (PARP) pathway [30,45]. Activation of PARP leads to the depletion of nicotinamide adenine dinucleotide (NAD+), which can contribute to changes in gene transcription and expression, further increase in generation of reactive oxygen species and divergence of glycolytic intermediates to other pathogenic pathways including aldose reductase and hexosamine pathways and formation of protein kinase C and advanced glycation endproducts [45,46]. Combined recurrent activation of these pathways overcomes the ability of defense mechanisms to repair the damage and eventually results in progressive nerve fiber dysfunction that is manifest as peripheral diabetic neuropathy.

In summary, our studies demonstrated that Mito-Q was effective in reducing multiple endpoints associated with neuropathy associated with an animal model of pre-diabetes as well as type 2 diabetes without significantly improving vascular function of epineurial arterioles suggesting that disruption of normal functionality of the mitochondria of neurons and perhaps Schwann cells contribute to the complex etiology of peripheral neuropathy.