Expression of hexokinase isoforms in the dorsal root ganglion of the adult rat and effect of experimental diabetes

Abstract

The effect of streptozotocin (STZ)-induced diabetes on expression and activity of hexokinase, the first enzyme and rate-limiting step in glycolysis, was studied in sensory neurons of lumbar dorsal root ganglia (DRG). The DRG and sciatic nerve of adult rats expressed the hexokinase I isoform only. Immunofluorescent staining of lumbar DRG demonstrated that small–medium neurons and satellite cells exhibited high levels of expression of hexokinase I. Large, mainly proprioceptive neurons, had very low or negative staining for hexokinase I. Intracellular localization and biochemical studies on intact DRG from adult rats and cultured adult rat sensory neurons revealed that hexokinase I was almost exclusively found in the mitochondrial compartment. Duration of STZ-diabetes of 6 or 12 weeks diminished hexokinase activity by 28% and 30%, respectively, in lumbar DRG compared with age matched controls (P < 0.05). Quantitative Western blotting showed no effect of diabetes on hexokinase I protein expression in homogenates or mitochondrial preparations from DRG. Immunofluorescent staining for hexokinase I showed no diabetes-dependent change in small–medium neuron expression in DRG, however, large neurons became positive for hexokinase I (P < 0.05). Such complex effects of diabetes on hexokinase I expression in the DRG may be due to glucose-driven up-regulation of expression or the result of impaired axonal transport and perikaryal accumulation in the large neuron sub-population. Because hexokinase is the rate-limiting enzyme of glycolysis these results imply that metabolic flux through the glycolytic pathway is reduced in diabetes. This finding, therefore, questions the role of high glucose-induced metabolic flux as a key driving force in reactive oxygen species generation by mitochondria.

Introduction

Diabetic sensory neuropathy in humans is associated with a spectrum of structural changes in peripheral nerve that includes axonal degeneration, microangiopathy, paranodal demyelination and loss of myelinated and unmyelinated fibers — the latter probably the result of a dying-back of distal axons (Malik et al., 2005, Yagihashi, 1997). In the streptozotocin (STZ)-diabetic rat and Bio-Breeding (BB) rat animal models of type I diabetes, similar structural abnormalities in peripheral nerve have been observed (Mizisin et al., 1999, Sima and Sugimoto, 1999, Yagihashi, 1997). The Diabetes Control and Complications Trial (DCCT) concluded that control of hyperglycemia is still the ideal means of preventing appearance of complications in diabetes, such as peripheral neuropathy (DCCT, 1990). However, such a goal remains an unrealized ideal and research continues to focus on biochemical transducers downstream from hyperglycemia that may directly induce neuropathic sensory nerve damage. Brownlee and colleagues have proposed that high cellular concentrations of glucose lead to mitochondrial dysfunction and the generation of damaging levels of reactive oxygen species (ROS) which then mediate cell degeneration (Nishikawa et al., 2000b). One component of such a theory presupposes that high glucose concentrations are converted by glycolysis into high levels of pyruvate that then enter the tricarboxylic acid cycle (TCA) within the mitochondrial matrix and provide high levels of reduced electron donors for electron transport. The inability of mitochondrial oxidative phosphorylation to efficiently deal with this high level of electron donation may lead to excessive superoxide production and oxidative stress (Nishikawa et al., 2000a, Nishikawa et al., 2000b).

Hexokinases serve as the gateway through which glucose enters glycolysis, by catalyzing the phosphorylation of glucose to yield glucose 6-phosphate (G-6-P), the initial and rate-limiting step of the glycolytic pathway (Wilson, 1995, Wilson, 2003). In mammals, four distinct hexokinase isozymes exist, designated types I, II, III and IV, with the latter commonly known as glucokinase (Wilson, 2003). The hexokinase I–III isozymes are 100 kDa molecules that display internal sequence repetition, and the N- and C-terminal halves have extensive sequence similarity to each other and to other members of the hexokinase family, in contrast, glucokinase has a molecular weight of approximately 50 kDa (Wilson, 1995). Based upon the kinetics of hexokinase isoform activities it is believed that hexokinase I functions primarily in a catabolic role, introducing glucose into glycolytic metabolism with the primary purpose of generating ATP. This is consistent with the ubiquitous expression of hexokinase I in tissues. In addition, hexokinase I is expressed at particularly high levels in the brain, which is virtually reliant on glycolytic metabolism of glucose to sustain a high rate of energy metabolism (Wilson, 2003). In contrast, hexokinase II is much more limited in its expression, primarily being found in insulin-sensitive tissues such as skeletal muscle and adipose tissue. Glucokinase has a relatively low affinity for glucose (K_m = 5–15 mM), with the result that the rate of glucose phosphorylation is responsive to physiologically relevant changes in plasma glucose concentration. Glucokinase is expressed primarily in the liver and pancreatic β-cells (Roncero et al., 2000).

Hexokinase I and II have the ability to bind specifically and reversibly to hydrophobic sequences at the outer mitochondrial membrane (Muller et al., 1994, Polakis and Wilson, 1985). A majority of the hexokinase activity is found in the mitochondrial fraction from homogenates of brain from various species (Kabir and Wilson, 1993). It has been demonstrated that hexokinase II interferes with the ability of bax to bind to mitochondria and release cytochrome c, consequently inhibiting bax-induced mitochondrial dysfunction and cell death (Pastorino et al., 2002). The physiological significance of this association of hexokinase with mitochondria, the primary site for oxidative metabolism, is that bound hexokinase has ‘preferential’ access to an intramitochondrial compartment of ATP. Subsequently, the rate of glucose phosphorylation (i.e. the rate at which glucose is being introduced into glycolysis) is directly correlated with the rate of mitochondrial oxidative phosphorylation (de Cerqueira Cesar and Wilson, 1995). Thus, binding of hexokinase to brain mitochondria provides a physical basis for coordination of the initial step of glycolytic metabolism occurring in the mitochondria.

Even though hexokinase has such a central role in glucose metabolism there is little information on its expression or activity in peripheral sensory neurons. Additionally, the effect of experimental diabetes on such parameters has not been investigated and possible role in sensory neuropathy studied. Therefore, this study describes the expression, localization and activity of the hexokinase isoforms in dorsal root ganglia (DRG) and sciatic nerve of adult rats. Furthermore, the effect of streptozotocin (STZ)-induced diabetes on expression, localization and activity of hexokinase was investigated. A central finding is that hexokinase activity is actually reduced in DRG of STZ-diabetic rats and would suggest that glycolytic delivery of substrates to the TCA cycle is, in fact, diminished. Therefore, the results question the validity of theories proposing that high glucose concentrations induce mitochondrial dysfunction under diabetic conditions in sensory neuron perikarya.