Oxidative stress has been implicated implicated in insulin resistance, and a new study by Hoehn et al. (1) adds some convincing evidence that one specific radical, superoxide generated in the mitochondria, may be a unifying cause. But the findings suggest that we may need to reconsider how we treat it.
The authors begin by suggesting that the failure of the recent ACCORD trial that found an increase in mortality when trying to lower glycated hemoglobin with drugs and/or insulin in diabetics may be explained by an increase in “intracellular stress by increasing nutrient delivery to an already stressed cell.” Note: to be objective, there are other theories on polypharmacy and the sort.
Here is a summary of their experiments:
Because an increase in mitochondrial oxidative stress has been found in differing models of inducing insulin resistance: inflammation, corticosteroids, chronic hyperinsulemia, or hyperlipidemia, it may be a common unifying mechanism governing it. In these 4 models, mitochondrial superoxide was found to be increased. Superoxide is formed mainly when electrons escape from complex I and/or III in the electron transport chain and react with oxygen.
Then, they induced insulin resistance in myocytes via incubation in the fatty acid palmitate and treated them with a mitochondrial uncoupler and an electron transport chain complex I inhibitor which reduce superoxide formation. Both resulted in a reversal of insulin resistance, suggesting superoxide as a causal mechanism for induction of insulin resistance.
Superoxide scavengers were tested next, with 3 mitochondrial superoxide dismutase mimetics which also reversed palmitate induced insulin resistance. A cytoplasmic superoxide dismutase mimetic had no effect, suggesting the mitochondrial derived superoxide is indeed the likely causal candidate.
Further evidence comes from the next experiment which involved genetic overexpression of the superoxide dismutase in the mitochondria (MnSOD); a 2.5 fold overexpression did not effect GLUT4 translocation in control myocytes but prevented insulin resistance induced by the 4 models described above, but seemed to potentiate insulin induced GLUT4 translocation, possibly because of an enhanced conversion of superoxide to hydrogen peroxide by superoxide dismutase.
Finally, using a compound to induce superoxide at complex III of the electron transport chain reduced insulin stimulated GLUT4 translocation in myocytes. To make sure it was the superoxide and not the inhibition of complex III which caused insulin resistance, the investigators used another compound that blocks the electron transfer to where the previous compound binds on complex III, which prevented insulin resistance. Two superoxide dismutase mimetics also prevented the complex III inhibition induced insulin resistance. Interestingly, the inhibition did not influence insulin signaling through Akt as its phosphorylation was normal at varying insulin concentrations.
The researchers then performed several in vivo studies in mice. Acute dosing of a mitochondrial superoxide dismutase mimetic with a high fat diet improved glucose tolerance by an increase in muscle and fat insulin sensitivity. Then, transgenic overexpression of mitochondrial superoxide dismutase showed a resistance to a insulin resistance induced by a high fat diet compared to controls after 1 week, 12 weeks, and 24 weeks. In the latter 2 periods, the transgenic mice gained the same amount of weight as the controls but were more resistant to insulin resistance. Then, in heterozygous mutant mice for mitochondrial superoxide dismutase, it was found that the resulting 70% reduction of the enzyme in muscle and fat impaired glucose tolerance on a low fat diet even with similar insulin concentrations to controls.
What does it mean?
Shuttling more nutrients through oxidative phosphorylation with insulin without increasing ATP consumption could increase mitochondrial superoxide because of ADP depletion and a reduction in electron carrier availability. In this manner, superoxide induced insulin resistance is an antioxidant in that it reduces further superoxide production, because it prevents too much glucose from entering the cell and undergoing oxidative phosphorylation. Superoxide also increases mitochondrial uncoupling and endogenous antioxidant genes in other studies, so insulin resistance seems to be another layer of protection that it adds to minimize cellular damage from overnutrition.
Fewer mitochondrial also may increase stress and result in an increased superoxide production. Lifestyle interventions such as intermittent fasting/fasting, calorie restriction, and endurance exercise promote mitochondrial biogenesis and are beneficial to diabetics in other research (especially exercise).
Interestingly, previous studies have implicated hydrogen peroxide as the cause of insulin resistance. Additionally, alpha lipoic acid, a compound available as a nutritional supplement, ameliorates insulin resistance and one of the proposed mechanisms is through its scavenging of hydrogen peroxide (and superoxide). Human studies also exist for the use of lipoic acid for diabetes treatment (example). I also recall that it stimulates mitochondrial biogenesis, likely via AMPK activation. This study paradoxically suggests that an increase in hydrogen peroxide may actually increase insulin sensitivity, which will require further contextual study. Though lipoic acid reduces glycation in other studies, the authors suggest (without comment on LA) that mitochondrial antioxidant therapy may theoretically impair protective mechanisms induced by superoxide without solving the underlying pathology, resulting in other problems such as AGE formation (glycation). Clearly this deserves further study, especially with needed alternative treatments with the ACCORD failure.
1. Hoehn KL, Salmon AB, Hohnen-Behrens C, Turner N, Hoy AJ, Maghzal GJ, Stocker R, Van Remmen H, Kraegen EW, Cooney GJ, Richardson AR, & James DE (2009). Insulin resistance is a cellular antioxidant defense mechanism. Proceedings of the National Academy of Sciences of the United States of America, 106 (42), 17787-92 PMID: 19805130