Early disruption of nerve mitochondrial and myelin lipid homeostasis in obesity-induced diabetes

Palavicini JP, Chen J, Wang C, Qin C, Baeuerle E, Wang X, Woo JA, Kang DE, Musi N, Dupree JL, and Han X. (2020) Early disruption of nerve mitochondrial and myelin lipid homeostasis in obesity-induced diabetes. JCI Insight, 5(21)

Diabetes mellitus is caused by the body’s inability to respond to or produce insulin, the hormone involved in regulating metabolism through encouraging the absorption of glucose. This dysregulation results in irregular breakdown of carbohydrates and can cause further complications including kidney damage, cardiovascular disease, blindness, and diabetic peripheral neuropathy. Diabetic peripheral neuropathy is a neurodegenerative disease affecting the peripheral nervous system causing a range of symptoms from pain and numbness to organ dysfunction. At this time treatment for neuropathy remains symptomatic with no effective therapies for halting progression. Hyperglycemia, obesity, insulin resistance, and hyperlipidemia are associated with the onset of diabetic peripheral neuropathy, and all are a part of metabolic syndrome. In addition, demyelination and mitochondrial dysfunction are known components of diabetic neuropathy. Lipids are a large component of both myelin and mitochondria both structurally and functionally, therefore Palavicini et al analyzed the role of hyperlipidemia in myelin using a diabetic mouse model.

Multidimensional mass spectrometry (MDMS) was used to analyze myelin lipid levels of the peripheral nervous system (PNS; dorsal root ganglia and sciatic nerves) and central nervous system (CNS; spinal cord and brainstem) tissues from the diabetic mice. To test for any changes in myelin, cerebroside and sulfatide, signature lipids in myelin, were examined. At one month the diabetic mice were visually obese with a 33% increased weight compared to the wildtype mice and a loss of myelin was detected in sciatic nerve tissue through reduced levels of sulfatide and cerebroside at 25% and 18%, respectively. Plasmalogen species 18:0/18:1 and 18:1/18:1 compose almost half of the phospholipid content in sciatic nerve tissue and were reduced by 27% and 12% in the diabetic mice. The spinal cord also showed sulfatide levels decrease by 18% and a 16% reduction of cerebroside. They went on to demonstrate that the effects were age dependent, with more severe reductions in sulfatide and cerebroside (33% and 29%, respectively) at two months of age. The lipidome in dorsal root ganglia, where neuronal cell bodies cluster, were tested but there was not a significant difference in the diabetic mice compared to the wild-type mice.

As myelin has a high lipid to protein ratio, it reasons that a change in lipid levels would result in functional or structural changes to myelin. In vivo electrophysiology was used to determine if any functional changes occurred in response to the reduction in myelin. At two months of age no differences in action potentials or conduction velocities were observed, however, by four months the diabetic mice showed reduced conduction velocities and a distinct reduction in motor action potential amplitude from 175mV in wildtype to ~45mV.

To further examine the structure of the myelin, electron microscopy (EM) was used. Consistent with the decrease in sulfatide, cerebroside, and plasmalogens, a decrease in myelin thickness was found in the PNS from 0.559μm in the wildtype mice at four months to 0.357μm in the diabetic mice. In addition, the g-ratio (diameter of the axon divided by the diameter of the myelinated fiber) of the sciatic nerve was significantly greater in the diabetic mice at 0.777 compared to 0.669 in the wildtype mice. Although no differences were seen within the PNS axon diameter between the two sets of mice, the observed functional difference suggests that other structural changes are likely present in the diabetic mice.

Palavicini et al were able to demonstrate a correlation between plasmalogen level and diabetic peripheral neuropathy progression in a mouse model of diabetes. These findings suggest that different types of diabetic neuropathy could require different types of treatment. Hyperglycemia is commonly thought to be the cause of this neuropathy, but hyperlipidemia could be another culprit. Interestingly, although alterations in myelin levels occurred early in the diabetic mice, the functional effects were not evident until later, observed through the reduced conduction velocities and decreased action potential amplitudes. This work supports the importance of adequate plasmalogen levels as an important determinant for the structure of myelin. As myelin is composed of roughly 70% phospholipids, most of which are plasmalogens, it is unsurprising that there was an association between plasmalogen reduction and improper neuronal function. Further work examining the relationship between plasmalogen levels and diabetes could provide a novel therapeutic option to treat this neuropathy.