Role of glucose regulatory mechanisms in diabetic retinopathy

Focusing on Sodium Glucose Cotransporter-2 and the Sympathetic Nervous System: Potential Impact in Diabetic Retinopathy

The prevalence of diabetes is at pandemic levels in today's society. Microvascular complications in organs including the eye are commonly observed in human diabetic subjects. Diabetic retinopathy (DR) is a prominent microvascular complication observed in many diabetics and is particularly debilitating as it may result in impaired or complete vision loss. In addition, DR is extremely costly for the patient and financially impacts the economy as a range of drug-related therapies and laser treatment may be essential. Prevention of microvascular complications is the major treatment goal of current therapeutic approaches; however, these therapies appear insufficient. Presently, sodium glucose cotransporter-2 (SGLT2) inhibitors may offer a novel therapy beyond simple glucose lowering. Excitingly, the EMPA-REG clinical trial, which focuses on the clinically used SGLT2 inhibitor empagliflozin, has been extremely fruitful and has highlighted beneficial cardiovascular and renal outcomes. The effects of SGLT2 inhibitors on DR are currently a topic of much research as outlined in the current review, but future studies are urgently needed to fully gain mechanistic insights. Here, we summarize current evidence and identify gaps that need to be addressed.

High glucose concentrations inhibit glucose phosphorylation, but not glucose transport, in human endothelial cells

Glucose uptake is autoregulated in a variety of cell types and it is thought that glucose transport is the major step that is subjected to control by sugar availability. Here, we examined the effect of high glucose concentrations on the rate of glucose uptake by human ECV-304 umbilical vein-derived endothelial cells. A rise in the glucose concentration in the medium led a dose-dependent decrease in the rate of 2-deoxyglucose uptake. The effect of high glucose was independent of protein synthesis and the time-course analysis indicated that it was relatively slow. The effect was not due to inhibition of glucose transport since neither the expression nor the subcellular distribution of the major glucose transporter GLUT1, nor the rate of 3-O-methylglucose uptake was affected. The total in vitro assayed hexokinase activity and the expression of hexokinase-I were similar in cells treated or not with high concentrations of glucose. In contrast, exposure of cells to a high glucose concentration caused a marked decrease in phosphorylated 2-deoxyglucose/free 2-deoxyglucose ratio. This suggests the existence of alterations in the rate of in vivo glucose phosphorylation in response to high glucose. In summary, we conclude that ECV304 human endothelial cells reduce glucose utilization in response to enhanced levels of glucose in the medium by inhibiting the rate of glucose phosphorylation, rather than by blocking glucose transport. This suggests a novel metabolic effect of high glucose on cellular glucose utilization.

Glut1 and glut3 expression, but not capillary density, is increased by cobalt chloride in rat cerebrum and retina

Treatment of rats with cobalt chloride [Co(II)], an agent that stimulates the expression of a set of hypoxia-responsive genes, for 10-12 days resulted in 1.45- and 1.40-fold increases in the content of Glut1 mRNA and Glut1 in cerebral gray matter, respectively (P<0. 05 for both changes). The increase in Glut1 content was associated with a significant increase in the content of Glut1 staining in microvessels isolated from cerebral gray matter, and in the intensity of Glut1 in microvessels of the frontal lobe and hippocampus assessed by immunohistochemistry. The abundance of Glut3 in cerebrum of Co(II)-treated rats also increased by 1.3-fold (P<0. 05), but the increase was not associated with a change in the content of Glut3 mRNA. In retina, treatment with Co(II) resulted in 2.48- and 1.23-fold increases in the content of Glut1 mRNA and Glut1 protein, respectively (P<0.05 for both changes); similar increases in Glut1 protein expression were observed in isolated retinal microvasculature. The content of Glut3 in retina also increased 1. 5-fold in Co(II)-treated rats (P<0.05). In addition, treatment with Co(II) resulted in a significant 2.2-fold increase in the expression of VEGF in the cerebrum. However, despite the Co(II)-induced increase in Glut1 expression in cerebral and retinal microvasculature and VEGF in cerebrum, there was no increase in the capillary density in either tissue. It is concluded that a 10-12 day exposure to Co(II), presumably acting through the hypoxia-signaling pathway, results in enhanced expression of both major glucose transporters in cerebral cortex and retina, without increasing the capillary density of either tissue.

Function of the N-acetyl-L-histidine system in the vertebrate eye. Evidence in support of a role as a molecular water pump

N-acetyl-L-histidine (NAH) is a major constituent of poikilotherm brain, eye, heart, and muscle, but for which there is no known function. NAH is characterized by high tissue concentrations, a high tissue/extracellular fluid (ECF) gradient, and by a continuous selective and regulated efflux into ECF. In the eye, there is a complete compartmentalization of the synthetic and hydrolytic enzymes, with synthesis of NAH from AcCoA and L-histidine (His) occurring in the lens, and its hydrolysis to acetate and His restricted to surrounding ocular fluids. Using 14C-isotopes, the cycling of NAH between lens and ocular fluids in a simple support medium consisting of NaCl (0.9%), Ca2+ (4 mEq/L) and D-glucose (5 mM) at pH 7.4 has previously been observed. In the present study, using the isolated lens of the goldfish eye, each of the components of that support medium has been individually varied in order to determine its effect on NAH release down its intercompartmental gradient. As a result of these and related studies, it is suggested that NAH may function as a metabolically recyclable gradient-driven molecular water pump. It is proposed that water influx or generation of metabolic water serves as the trigger mechanism to open a Ca-dependent gate for the release of NAH down its gradient, along with its associated water. Preliminary analyses suggest that in addition to its potential for multiple daily cycles, a strongly ionized hydrophilic molecule, such as NAH, may include a large power function as a result of its attraction to water, and it has been calculated that an aqua complex of each NAH molecule may have 33 dipole-dipole-associated water molecules as it passes into ECF. It is this unique combination of a capacity for multiple cycles per day, coupled with a large power function, that may allow for such an intracellular osmolyte to be present in relatively low concentration in comparison to total cellular osmolality, and yet to perform a large and important task with little expenditure of energy. With each NAH molecule recycled up to 10 times/d, and a power factor of 33, there could be 330 mmol of water transported/mmol of NAH each day. With typical NAH concentrations in brains of poikilothermic vertebrates of 5-10 mmol/kg, there is the potential for up to 3.3 mol (60 mL) of water to be removed each day/kg of brain, a value that represents about 8% of total brain water content. Dewatering of the released osmolyte would occur in two additional steps, consisting of its hydrolysis and the subsequent active uptake of its metabolites. It is also suggested that NAH is the archetype of several metabolically and structurally related cellular osmolytes found in both poikilotherms and homeotherms, for which there is similarly no known function, and these may form a family of cycling hydrophilic osmolytes that serve as molecular water pumps in a variety of tissues. These include the basic His containing derivatives: NAH, carnosine, anserine, ophidine, and homocarnosine, and the acidic aspartate derivatives: N-acetyl-L-aspartate (NAA) and N-acetyl-L-aspartylglutamate (NAAG). In each of these cases, the high intracellular/extracellular osmolyte gradient appears to be maintained by combining a hydrophilic protein amino acid with a nonprotein moiety to block its use in other intracellular metabolic pathways, and by blocking catabolism of the derivative by maintaining its hydrolytic enzyme in an extracytosolic membrane or extracellular compartment. Unlike other known water-regulating mechanisms, the proposed cellular system is unique in that as a water pump, it can function as a water regulator independently of extracellular solute composition or osmolality. Finally, based on the hypothesis developed, the NAH system would represent the first cellular water pump to be identified.

Regulation of glucose transporters (GLUT-1 and GLUT-3) in human retinal endothelial cells

The regulation of glucose transporters (GLUT-1 and GLUT-3), in terms of both mRNA and protein, in human retinal endothelial cells was investigated. The cells responded within 1 h of exposure to 5 mM glucose with an increase in the level of GLUT-3 mRNA that was due to an increase in the transcription of the 4.1 kb mRNA of the gene for GLUT-3. In the absence of glucose, the gene for GLUT-1 was not transcribed but the level of GLUT-3 mRNA was increased in these conditions and this was the result of an increase in the transcription of the 4.1 kb mRNA. The level of GLUT-1 and GLUT-3 mRNA was maximal when the cells were exposed to 15 mM glucose. These results are discussed in the light of the glucose regulatory potential of the retinal microvasculature and the implications that this may have for the mechanisms of diabetic retinopathy.