- Email: [email protected]
Recent advances in the therapy of diabetic peripheral neuropathy by means of an aldose reductase inhibitor
Recent Advances in the Therapy of Diabetic Peripheral Neuropathy by Means of an Aldose Reductase Inhibitor
Nerve conduction slowing, a hallmark of both experimental and human diabetic neuropathy, is improved or corrected by administration of aldose reductase inhibitors such as sorbinil. Recent experiments in animals attribute acutely reversible nerve conduction slowing in diabetes to a myo-inositol-related defect in nerve sodium-potassium adenosinetriphosphatase, which generates the transmembrane sodium and potassium potentials necessary for nerve impulse conduction and the sodium gradient necessary for sodium-dependent uptake of substrates. This myo-inositol-related abnormality in sodium-potassium adenosinetriphosphatase function is currently viewed as a cyclic metabolic defect involving sequential alteration of sodium-dependent myo-inositol uptake, myo-inositol content, myo-inositol incorporation into membrane phospholipids, and phospholipid-dependent sodium-potassium adenosinetriphosphatase function in peripheral nerve. Aldose reductase inhibitors have been shown to normalize both nerve myoinositol content and nerve sodium-potassium adenosinetriphosphatase activity. These observations suggest that the acute effects of aldose reductase inhibitors on nerve conduction in both animals and humans with diabetes may be mediated by correction of an underlying myo-inositol-related nerve sodium-potassium adenosinetriphosphatase defect. Furthermore, this sorbinil-corrected sodium-potassium adenosinetriphosphatase defect in diabetic nerve may contribute to other biochemical, functional, and structural abnormalities present in diabetic peripheral neuropathy.
DOUGLAS A. GREENE, M.D. SARAH A. LATTIMER, B.S. Pittsburgh,
Pennsylvania
From the Diabetes Research Laboratories of the Department of Medicine, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania. Requests for reprints should be addressed to Dr. Douglas A. Greene, Room 3304, Presbyterian University Hospital, 230 Lothrop Street, Pittsburgh, Pennsylvania 15261.
November
Diabetes-associated complications (macroangiopathy, nephropathy, neuropathy, and retinopathy) are heterogeneous disorders that are probably due to a combination of causes. Nevertheless, for some time, ‘it has been thought that the altered metabolic environment in diabetes mellitus exerts a conditioning influence on the development of these complications. Recent animal and\ in vitro studies have identified several interrelated metabolic abnormalities in nerve, glomerulus, retina, and arterial wall that are attributable to the elevated glucose concentrations. In particular, increased activity of the sorbitol pathway and decreased myo-inositol content seem to become manifest in these tissues as a reduction in sodium-potassium adenosinetriphosphatase activity occurs [l-6]. This enzyme is crucial to the generation of the electrochemical potential necessary for nerve conduction, as well as for a variety of other biochemical and biophysical functions, including water and electrolyte homeostasis and the active concentration of metabolic substrates. These metabolic changes may, in combination, induce a variety of biochemical and bio-
15, 1985
The
American
Journal
of Medicine
Volume
79
(suppl 5A)
13
SYMPOSIUM
TABLE
ON ALDOSE
I
REDUCTASE
Parallel Responses Activity, and Motor
INHIBITION-GREENE
and LATTIMER
of Rat Nerve Myo-lnositol Content, Sodium-Potassium Adenosinetriphosphatase Nerve Conduction Velocity to Hyperglycemia and Metabolic Intervention* Motor
Plasma Glucose (mg/lOlJ ml)
Status
of Animal Nondiabetic Diabetic Diabetic + insulin replacement Diabetic + myo-inositol supplementation Diabetic + sorbinil treatment *Data
from
Plasma Myo-lnositol (ccmol/liter)
178i7 566 i 30 75+- 18
600~~
623 f 25
Sodium-Potassium Adenosinetriphosphatase (mmol/kg . h)
2 3 2
3.0 2 0.2 2.2 i 0.1 3.0 i 0.2
62 i 1 51 f 1 63 f 1
97 5 5 55 i- 3 120 + 7
245 + 27
3.9 i 0.2
61 i
3.1 f 0.1
-
305
3
1
88 IL 4
962
5
[I].
physical alterations that are highly relevant to the pathogenesis of diabetes-associated complications. Some of the changes that occur in tissues susceptible to the complications of diabetes may be related to alterations in the metabolism of myo-inositol, a polyol that is structurally related to glucose but is different biochemically. The major relevance of myo-inositol appears to be in its reversible incorporation into membrane phospholipids, and it seems to play an important role in the function of membranes. The concentration of myo-inositol in peripheral nerve, glomerulus, and retina is reduced in acute experimental diabetes, but is restored to normal by administration of insulin. Furthermore, there appears to be a correlation between the function of peripheral nerve and retina and their myo-inositol content. THE POLYOL
(SORBITOL)
PATHWAY
Increased sorbitol metabolism resulting from hyperglycemia and leading to abnormal accumulation of sorbitol and fructose in peripheral nerve has been proposed as a pathogenetic factor in diabetic neuropathy [l]. The sequential reduction and oxidation of glucose to sorbitol and sorbitol to fructose by the enzymes aldose reductase and sorbitol dehydrogenase are regulated primarily by the intracellular concentrations of glucose and sorbitol, respectively. Hence, these two linked reactions, which comprise the sorbitol or polyol pathway, are accelerated by elevated glucose concentrations in peripheral nerve, glomerulus, retina, and arterial wall [l-6]. Osmotic swelling due to the intracellular accumulation of sorbitol has been proposed as a cause of both functional and morphologic alterations in diabetic nerves. This “osmotic hypothesis” is derived primarily from studies of experimental sugar cataracts in rats in which experimental diabetes or glucose, galactose, or xylose intoxication is accompanied by millimolar accumulations of polyol,
14
Nerve
Conduction Velocity (m/set)
35k 35i 30+
32
Nerve
Nerve Myo-lnositol (mmol/kg)
November
15, 1985
The American
Journal
of Medicine
marked increases in ocular lens water content, and rapid development of lenticular opacities. Extrapolation of this hypothesis to diabetic peripheral nerve, retina, glomeruIus, and arterial wall is, unfortunately, beset by several conceptual obstacles. For instance, sorbitol accumulates only in micromolar rather than millimolar concentrations, reducing the likelihood of significant osmotic effects (unless the accumulation is highly localized anatomically) [I -
‘31. Levels of glucose, sorbitol, and fructose in peripheral nerves are elevated in both human diabetes [7] and experimental diabetes [I]. Specific aldose reductase inhibitors improve nerve conduction velocities in patients with diabetes [8,9], although their short-term effects in humans are small. As polyol-induced osmotic effects are unlikely to make a major contribution to the slowing of conduction in the diabetic peripheral nerve, other manifestations of increased polyol pathway activity must be evoked to explain the beneficial effects of aldose reductase inhibitors on nerve function. SLOWING
OF NERVE CONDUCTION
Relationship to Sodium-Potassium AdenosinetriphosphataseActivityandMyo-InositolMetabolism. Reversible slowing of nerve conduction in animals with diabetes is a consequence of acute insulin deficiency [I O-l 21 and occurs in the absence of segmental demyelination or axonal degeneration [11,13,14]. It has been attributed to acute metabolic disturbances in axons or Schwann cells or their endoneurial environment [l]. Although a variety of metabolic defects have been identified in peripheral nerves of animals with acute diabetes, to date only one sequence--that involving altered neural myo-inositol and sodium-potassium adenosinetriphosphatase-is directly and consistently implicated in the slowing of nerve conduction [1,7,10,15-171. This conclusion is based on the
Volume
79 (suppl
5A)
SYMPOSIUM
REGULATION
INHIBITION-GREENE
and LATTIMER
OF INTRACELLULAR
MYO-INOSITOL
In peripheral nerve, there is normally a tissue-to-plasma myo-inositol concentration gradient of 1 to 90 to 100, but this gradient is reduced in both human diabetes [S] and animal diabetes [10,18,22,24]. A specific, high-affinity, sodium-dependent, carrier-mediated myo-inositol transport system probably contributes to the establishment and/or maintenance of the concentration gradient in nerve [25]. Elevated concentrations of glucose competitively inhibit the uptake of myo-inositol through the system [25], providing a potential mechanism by which the myo-inositol content of nerve is reduced in diabetes mellitus. The low transport capacity and high concentration gradient of this transport system suggest that factors influencing passive efflux of myo-inositol might also influence the concentration gradient. A similar transport system is presumed to be present in glomerulus and retina.
LINK BETWEEN MYO-INOSITOL METABOLISM AND SODIUM-POTASSIUM ADENOSINETRIPHOSPHATASE In mammalian cells, membrane-bound sodium-potassium adenosinetriphosphatase is regulated by a variety of intrinsic intracellular modulators including cytoplasmic magnesium, calcium, adenosine triphosphate, and pH; the characteristics of the surrounding plasma membrane; and cytoplasmic and extracellular concentrations of sodium and potassium [20]. In some tissues, additional extrinsic (for example, hormonal) modulation may occur via phosphorylation and dephosphorylation of either sodium-potassium adenosinetriphosphatase itself [20] or an associated membrane protein [21], or by regulation of
15, 1985
REDUCTASE
sodium-potassium adenosinetriphosphatase protein synthesis or degradation [20]. However, the sodium-potassium adenosinetriphosphatase defect in diabetic peripheral nerve is not influenced acutely by insulin in vitro [22] and is expressed in broken-cell nerve homogenates in which the acute effects of water-soluble modulators have been eliminated [1,23]. Therefore, the defect appears intrinsic to the sodium-potassium adenosinetriphosphatase-membrane complex. The defect is entirely prevented by in vivo supplementation with myo-inositol, which has been shown in previous studies in animal models of human diabetic neuropathy to prevent the characteristic decreases in myo-inositol levels in nerve and motor conduction velocity (Table I). Thus, it is concluded that the sodium-potassium adenosinetriphosphatase defect reflects some as yet unidentified disruption in inositol phospholipid metabolism in diabetic nerve [2]. Recent evidence suggests that metabolites of inositol phospholipid catabolism stimulate the activity of a cyclic adenosine monophosphate-independent, phospholipid-dependent and calcium-dependent protein kinase (protein kinase C) that activates sodium-potassium adenosinetriphosphatase, and that a deficiency in myo-inositol interferes with this sequence, producing a sodium-potassium adenosinetriphosphatase defect in diabetic nerve.
observed parallel responses of nerve myo-inositol content, sodium-potassium adenosinetriphosphatase activity, and motor nerve conduction velocity in animals with acute diabetes during insulin replacement or oral myo-inositol supplementation, in which parallelism persists despite continuing hyperglycemia and elevated concentrations of glucose, sorbitol, and fructose in nerve (Table I) [l ,10,16,17]. Thus, impairment of nerve sodium-potassium adenosinetriphosphatase activity and nerve conduction in animals with acute diabetes is a consequence of altered myo-inositol metabolism in the diabetic nerve, which secondarily alters the structure and/or function of the sodium-potassium adenosinetriphosphatase-membrane complex [18]. ,pimilar relationships between tissue sodium-potassium myo-inositol : metabolism and adenosinetriphosihatase activity have recently been reported for diabetic renal gibmerulus [2,3] and retina [4,5]. Relationship between Aldose Reductase Inhibitors and Myo-lnositol Metabolism. Aldose reductase inhibitors such as sorbinil completely prevent the decrease in myo-inositol content and altered sodium-potassium adenosinetriphosphatase activity in diabetic peripheral nerve and glomerulus [l-4]. As mentioned previously, the administration of myo-inositol to correct characteristically reduced tissue content without influencing the concentration of sorbitol in the nerve and glomerulus restores conduction velocity in diabetic nerve and increased urinary protein loss toward normal values [l-4,19]. Thus, increased polyol pathway activity in diabetes appears to contribute to the disruption in myo-inositol metabolism. Alternatively, aldose reductase inhibitors influence the metabolism of myo-inositol through a mechanism other than aldose reductase inhibition, and their action in diabetic tissue is attributable, at least in part, to their effects on the myo-inositol-related defect in sodium-potassium adenosinetriphosphatase discussed earlier. Hence, the two most commonly cited biochemical abnormalities in tissues susceptible to diabetic complications-altered myoinositol and sorbitol metabolism-are closely intertwined both metabolically and functionally.
November
ON ALDOSE
NERVE MYO-INOSITOL TRANSPORT AND SODIUMPOTASSIUM ADENOSINETRIPHOSPHATASE The peripheral nerve myo-inositol transport system is activated by the sodium gradient generated by sodium-potassium adenosinetriphosphatase [25]. Therefore, it might be expected that the reduced sodium-potassium adenosinetriphosphatase activity of diabetic peripheral nerve would secondarily impair the sodium-dependent uptake of myo-inositol. Recent experiments confirm that this is so and that sodium-dependent myo-inositol uptake remains
The
American
Journal
of Medicine
Volume
79 (suppl
5A)
15
SYMPOSIUM
ON ALDOSE REDUCTASE
INHIBITION-GREENE
and LAlTlMER
Hyperglycemia I
-4
Compititive inhibition
myo-inositol
fPolyol oathwav
uptake
P 4NalK ATPase activity
4Tissue myo-inositol &Membrane phosphoinositide
Figure 1. Proposed cyclic metabolic defect involving myoinositol, phosphoinositide, and the sodium-potassium adenosinetriphosphatase that is induced by hyperglycemia in peripheral nerve.
reduced in diabetic peripheral nerve in vitro even after tissue glucose concentrations are normalized. Furthermore, this impairment correlates with the reduction in sodiumpotassium adenosinetriphosphatase activity under various experimental conditions [3], suggesting that reduced sodium-potassium adenosinetriphosphatase activity in diabetic nerve further compromises endoneurial sodiumdependent uptake of myo-inositol. Thus, these relationships set the stage for a self-sustaining cyclic metabolic disruption involving myo-inositol and sodium-potassium adenosinetriphosphatase in diabetic peripheral nerve (Figure 1). A similar sequence may also exist in other tissues susceptible to diabetic complications. UNIFYING
METABOLIC
HYPOTHESIS
Hyperglycemia alters the metabolism in nerves in several ways. Competitive inhibition of sodium-dependent uptake of myo-inositol and/or increased polyol (sorbitol) pathway activity reduce the myo-inositol content of nerve, alter the phosphoinositide metabolism of nerve, diminish protein kihase C activity [26], and impair sodium-potassium adenosinetriphosphatase activity, further reducing sodium-gradient-dependent myo-inositol uptake and there-
by initiating a self-sustaining cycle. Impairment of the electrogenic sodium-potassium adenosinetriphosphatase activity acutely and secondarily reduces nerve conduction velocity [27] while probably also inhibiting other sodiumgradient-dependent processes such as amino a&d uptake and intracellular water homeostasis. Other defects may also occur as a result of alterations in the activity of inositol phospholipid metabolism or protein kinase C that are independent of sodium-potassium adenosinetriphosphatase and that may possibly involve other membranebound proteins, such as the voltage-dependent sodiumchannel or other membrane cationic adenosinetriphosphatases. These abnormalities may become self-sustaining cyclic defects, with potentially widespread pathophysiolbgic implications, which lead to impaired nerve conduction, axonal transport, intermediary-metabolism, and water balance, and possibly also contribute to the later structural defects in diabetic peripheral nerve. By analogy, a similar series of events may occur in other tissues in which the myo-inositol conteht is reduced through exposure to elevated concentrations of glucose. LINK BETWEEN P~LY~L METABOLISM AND THE MYO-IfiOSlTOL-RELATED SODIUkl-POTASSIUM ADENOSINETRIPHOSPHATASE DEFECT
Increased polyol pathway activity and the effects of aldose reductase inhibitors on peripheral nerve, renal glomerulus, retina, and arterial wall may be mediated primarily through the my&inositol-related sodium-potassium adenosinetriphosphatase defect described earlier. Elucidation of the link between sorbitol metabolism and myo-inositol or sodium-potassium adenosinetriphdsphatase is, therefoie, essential for comprehending the action -of aldose reductase inhibitors. Aldose reductase inhibition would be expected to alter the intracellular concentrations of several polyol pathway intermediates including sorbitol, fructose, oxidized nicotinamide-adenine dinucleotide phosphate, reduced nicotinamide-adenine dinucleotide phosphate, oxidized nicdtinamide-adenine dinucleotide, and reduced nicotinamide-adenine dinucleotide. Possible links between these intermediates and myo-inositol and sodium-potassium adenosinetriphosphatase metabolism remain to be elucidated.
REFERENCES 1.
2.
3.
16
Greene DA, Lattimer SA, Ulbrecht JS, Carroll PB: Glucose-induced alterations in nerve metabolism: current perspective on the pathogenesis of diabetic neuropathy and future directions for research and therapy. Diabetes Care 1985; 8: 290-299. Beyer-Mears A, Ku L, Cohen MP: Glomerular polyol accumulation in diabetes and its prevention by oral sorbinil. Diabetes 1984; 33: 604-607. Cohen MP: Reduced glomerular sodium-potassium
November 15, 1985
The American Journal of Medicine
4.
5. 6.
adenosinetriphosphatase activity in acute streptozotocin diabetes and its prevention by oral sorbinil. Diabetes (in press). MacGregor LC, Rosecan LR, Laties AM, Matchinsky FM: Microanalysis of total lipid, glucose, sorbitol, and myo-inositol in individual retinal layers of normal and alloxan diabetic rabbits (abstr). Diabetes 1984; 33(suppl 1): 89A. MacGregor L, Matchinsky F: Personal communication. Morrison AD: Aortic smooth muscle metabolism: effect of polyol
Volume 79 (suppl 5A)
SYMPOSIUM
7.
8.
9.
10.
11.
12. 13. 14.
15.
16.
17.
pathway inhibition (abstr). Clin Res 1984; 32(suppl 5): 851A. Mayhew JA, Gillon KRW, Hawthorne JN: Free and lipid inositol, sorbltol, and sugars in sciatic nerve obtained post mortem from diabetic and control subjects. Diabetologia 1983; 24: 1315. Fagius J, Jameson S: Effects of aldose reductase inhibitor treatment in diabetic polyneuropathy-a clinical and neurophysiological study. J Neurol Neurosurg Psychiatry 1981; 44: 9911001. Judzewitsch RG, Jaspan JB, Polonsky KS, et al: Aldose reductase inhibition improves nerve conduction velocity in diabetic patients. N Engl J Med 1983; 308: 119-125. Greene DA, DeJesus PV, Winegrad Al: Effects of insulin and dietary myo-inositol on impaired peripheral motor nerve conduction velocity in, acute streptozotocin diabetes. J Clin Invest 1975; 55: 1326-1336. Jakobsen J: Early and preventable changes of peripheral nerve structure and function in insulin-deficient diabetic rats. J Neural Neurosurg Psychiatry 1979; 42: 502-518. Sima AAF: Peripheral neuropathy in the spontaneously diabetic BB-Wistar rat. Acta Neuropathol (Ben) 1980; 51: 223-227. Sharma AK, Thomas PK: Peripheral nerve structure and function in experimental diabetes. J Neurol Sci 1974; 23: l-15. Brown MJ, Sumner AJ, Greene DA, et al: Distal neuropathy in experimental diabetes mellitus. Ann Neurol 1980; 8: 168178. Greene DA: Metabolic abnormalities in diabetic peripheral nerve: relation to impaired function. Metabolism 1983; 32 (suppl 1): 118-l 23. Greene DA, Lewis RA, Lattimer SA, Brown MJ: Selective effects of myo-inositol administration on sciatic and tibia1 motor nerve conduction parameters in the streptozotocin-diabetic rat. Diabetes 1982; 31: 573-578. Mayer JH, Tomlinson DR: Prevention of defects of axonal trans-
November
15, 1985
18.
19.
20.
21,
22.
23.
24.
25.
26.
27.
The
ON ALDOSE
REDUCTASE
INHIBITION-GREENE
and LATTIMER
port and nerve conduction velocity by oral administration of myo-inositol or an aldose reductase inhibitor in streptozotocin-diabetic rats. Diabetologia 1983; 25: 433-438. Greene DA, Yagihashi S, Lattimer SA, Sima AAF: Nerve Na+K+ATPase, conduction, and myo-inositol in the insulindeficient BB rat. Am J Physiol 1984; 247: E534-E539. Beyer-Mears A, Ku L, Cohen MP: Glomerular polyol accumulation in diabetes and its prevention with sorbinil (abstr). Diabetes 1984; 33 (suppl 1): 89A. Trachtenberg MC, Packey DJ, Sweeney T: In vivo functioning of the NA+K+-activated ATPase. Curr Top Cell Regul 1981; 19: 159-217. Lingham RB, Sen AK: Regulation of rat brain (Na+K+)-ATPase activity by cyclic AMP. Biochim Biophys Acta 1982; 688: 475485. Greene DA, Winegrad Al: Effects of acute experimental diabetes on composite energy metabolism in peripheral nerve axons and Schwann cells. Diabetes 1981; 30: 967-974. Das PK, Bray G, Aguayo AJ, Rasminsky M: Diminished ouabain-sensitive sodium-potassium ATPase activity in sciatic nerves of rats with streptozotocin-induced diabetes. Exp Neural 1976; 53: 285-288. Palmano KP, Whiting PH, Hawthorne JN: Free and lipid myoinositol in tissues from rats with acute and less severe streptozotocin-induced diabetes. Biochem J 1977; 167: 229-235. Greene DA, Lattimer SA: Sodium and energy dependent uptake of myo-inositol by rabbit peripheral nerve. J Clin Invest 1972; 70: 1009-1018. Greene DA, Lattimer SA: Phorbol ester corrects the myo-inositol-related Na/K-ATPase defect in diabetic nerve (abstr). Clin Res 1985; 33: 569A. Brismar T, Sima AAF: Changes in nodal function in nerve fibres of the spontaneously diabetic BB-Wistar rat: potential clamp analysis. Acta Physiol Stand 1981; 113: 499-506.
American
Journal
of Medicine
Volume
79 (suppl
5A)
17
Nerve conduction slowing, a hallmark of both experimental and human diabetic neuropathy, is improved or corrected by administration of aldose reductase inhibitors such as sorbinil. Recent experiments in animals attribute acutely reversible nerve conduction slowing in diabetes to a myo-inositol-related defect in nerve sodium-potassium adenosinetriphosphatase, which generates the transmembrane sodium and potassium potentials necessary for nerve impulse conduction and the sodium gradient necessary for sodium-dependent uptake of substrates. This myo-inositol-related abnormality in sodium-potassium adenosinetriphosphatase function is currently viewed as a cyclic metabolic defect involving sequential alteration of sodium-dependent myo-inositol uptake, myo-inositol content, myo-inositol incorporation into membrane phospholipids, and phospholipid-dependent sodium-potassium adenosinetriphosphatase function in peripheral nerve. Aldose reductase inhibitors have been shown to normalize both nerve myoinositol content and nerve sodium-potassium adenosinetriphosphatase activity. These observations suggest that the acute effects of aldose reductase inhibitors on nerve conduction in both animals and humans with diabetes may be mediated by correction of an underlying myo-inositol-related nerve sodium-potassium adenosinetriphosphatase defect. Furthermore, this sorbinil-corrected sodium-potassium adenosinetriphosphatase defect in diabetic nerve may contribute to other biochemical, functional, and structural abnormalities present in diabetic peripheral neuropathy.
DOUGLAS A. GREENE, M.D. SARAH A. LATTIMER, B.S. Pittsburgh,
Pennsylvania
From the Diabetes Research Laboratories of the Department of Medicine, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania. Requests for reprints should be addressed to Dr. Douglas A. Greene, Room 3304, Presbyterian University Hospital, 230 Lothrop Street, Pittsburgh, Pennsylvania 15261.
November
Diabetes-associated complications (macroangiopathy, nephropathy, neuropathy, and retinopathy) are heterogeneous disorders that are probably due to a combination of causes. Nevertheless, for some time, ‘it has been thought that the altered metabolic environment in diabetes mellitus exerts a conditioning influence on the development of these complications. Recent animal and\ in vitro studies have identified several interrelated metabolic abnormalities in nerve, glomerulus, retina, and arterial wall that are attributable to the elevated glucose concentrations. In particular, increased activity of the sorbitol pathway and decreased myo-inositol content seem to become manifest in these tissues as a reduction in sodium-potassium adenosinetriphosphatase activity occurs [l-6]. This enzyme is crucial to the generation of the electrochemical potential necessary for nerve conduction, as well as for a variety of other biochemical and biophysical functions, including water and electrolyte homeostasis and the active concentration of metabolic substrates. These metabolic changes may, in combination, induce a variety of biochemical and bio-
15, 1985
The
American
Journal
of Medicine
Volume
79
(suppl 5A)
13
SYMPOSIUM
TABLE
ON ALDOSE
I
REDUCTASE
Parallel Responses Activity, and Motor
INHIBITION-GREENE
and LATTIMER
of Rat Nerve Myo-lnositol Content, Sodium-Potassium Adenosinetriphosphatase Nerve Conduction Velocity to Hyperglycemia and Metabolic Intervention* Motor
Plasma Glucose (mg/lOlJ ml)
Status
of Animal Nondiabetic Diabetic Diabetic + insulin replacement Diabetic + myo-inositol supplementation Diabetic + sorbinil treatment *Data
from
Plasma Myo-lnositol (ccmol/liter)
178i7 566 i 30 75+- 18
600~~
623 f 25
Sodium-Potassium Adenosinetriphosphatase (mmol/kg . h)
2 3 2
3.0 2 0.2 2.2 i 0.1 3.0 i 0.2
62 i 1 51 f 1 63 f 1
97 5 5 55 i- 3 120 + 7
245 + 27
3.9 i 0.2
61 i
3.1 f 0.1
-
305
3
1
88 IL 4
962
5
[I].
physical alterations that are highly relevant to the pathogenesis of diabetes-associated complications. Some of the changes that occur in tissues susceptible to the complications of diabetes may be related to alterations in the metabolism of myo-inositol, a polyol that is structurally related to glucose but is different biochemically. The major relevance of myo-inositol appears to be in its reversible incorporation into membrane phospholipids, and it seems to play an important role in the function of membranes. The concentration of myo-inositol in peripheral nerve, glomerulus, and retina is reduced in acute experimental diabetes, but is restored to normal by administration of insulin. Furthermore, there appears to be a correlation between the function of peripheral nerve and retina and their myo-inositol content. THE POLYOL
(SORBITOL)
PATHWAY
Increased sorbitol metabolism resulting from hyperglycemia and leading to abnormal accumulation of sorbitol and fructose in peripheral nerve has been proposed as a pathogenetic factor in diabetic neuropathy [l]. The sequential reduction and oxidation of glucose to sorbitol and sorbitol to fructose by the enzymes aldose reductase and sorbitol dehydrogenase are regulated primarily by the intracellular concentrations of glucose and sorbitol, respectively. Hence, these two linked reactions, which comprise the sorbitol or polyol pathway, are accelerated by elevated glucose concentrations in peripheral nerve, glomerulus, retina, and arterial wall [l-6]. Osmotic swelling due to the intracellular accumulation of sorbitol has been proposed as a cause of both functional and morphologic alterations in diabetic nerves. This “osmotic hypothesis” is derived primarily from studies of experimental sugar cataracts in rats in which experimental diabetes or glucose, galactose, or xylose intoxication is accompanied by millimolar accumulations of polyol,
14
Nerve
Conduction Velocity (m/set)
35k 35i 30+
32
Nerve
Nerve Myo-lnositol (mmol/kg)
November
15, 1985
The American
Journal
of Medicine
marked increases in ocular lens water content, and rapid development of lenticular opacities. Extrapolation of this hypothesis to diabetic peripheral nerve, retina, glomeruIus, and arterial wall is, unfortunately, beset by several conceptual obstacles. For instance, sorbitol accumulates only in micromolar rather than millimolar concentrations, reducing the likelihood of significant osmotic effects (unless the accumulation is highly localized anatomically) [I -
‘31. Levels of glucose, sorbitol, and fructose in peripheral nerves are elevated in both human diabetes [7] and experimental diabetes [I]. Specific aldose reductase inhibitors improve nerve conduction velocities in patients with diabetes [8,9], although their short-term effects in humans are small. As polyol-induced osmotic effects are unlikely to make a major contribution to the slowing of conduction in the diabetic peripheral nerve, other manifestations of increased polyol pathway activity must be evoked to explain the beneficial effects of aldose reductase inhibitors on nerve function. SLOWING
OF NERVE CONDUCTION
Relationship to Sodium-Potassium AdenosinetriphosphataseActivityandMyo-InositolMetabolism. Reversible slowing of nerve conduction in animals with diabetes is a consequence of acute insulin deficiency [I O-l 21 and occurs in the absence of segmental demyelination or axonal degeneration [11,13,14]. It has been attributed to acute metabolic disturbances in axons or Schwann cells or their endoneurial environment [l]. Although a variety of metabolic defects have been identified in peripheral nerves of animals with acute diabetes, to date only one sequence--that involving altered neural myo-inositol and sodium-potassium adenosinetriphosphatase-is directly and consistently implicated in the slowing of nerve conduction [1,7,10,15-171. This conclusion is based on the
Volume
79 (suppl
5A)
SYMPOSIUM
REGULATION
INHIBITION-GREENE
and LATTIMER
OF INTRACELLULAR
MYO-INOSITOL
In peripheral nerve, there is normally a tissue-to-plasma myo-inositol concentration gradient of 1 to 90 to 100, but this gradient is reduced in both human diabetes [S] and animal diabetes [10,18,22,24]. A specific, high-affinity, sodium-dependent, carrier-mediated myo-inositol transport system probably contributes to the establishment and/or maintenance of the concentration gradient in nerve [25]. Elevated concentrations of glucose competitively inhibit the uptake of myo-inositol through the system [25], providing a potential mechanism by which the myo-inositol content of nerve is reduced in diabetes mellitus. The low transport capacity and high concentration gradient of this transport system suggest that factors influencing passive efflux of myo-inositol might also influence the concentration gradient. A similar transport system is presumed to be present in glomerulus and retina.
LINK BETWEEN MYO-INOSITOL METABOLISM AND SODIUM-POTASSIUM ADENOSINETRIPHOSPHATASE In mammalian cells, membrane-bound sodium-potassium adenosinetriphosphatase is regulated by a variety of intrinsic intracellular modulators including cytoplasmic magnesium, calcium, adenosine triphosphate, and pH; the characteristics of the surrounding plasma membrane; and cytoplasmic and extracellular concentrations of sodium and potassium [20]. In some tissues, additional extrinsic (for example, hormonal) modulation may occur via phosphorylation and dephosphorylation of either sodium-potassium adenosinetriphosphatase itself [20] or an associated membrane protein [21], or by regulation of
15, 1985
REDUCTASE
sodium-potassium adenosinetriphosphatase protein synthesis or degradation [20]. However, the sodium-potassium adenosinetriphosphatase defect in diabetic peripheral nerve is not influenced acutely by insulin in vitro [22] and is expressed in broken-cell nerve homogenates in which the acute effects of water-soluble modulators have been eliminated [1,23]. Therefore, the defect appears intrinsic to the sodium-potassium adenosinetriphosphatase-membrane complex. The defect is entirely prevented by in vivo supplementation with myo-inositol, which has been shown in previous studies in animal models of human diabetic neuropathy to prevent the characteristic decreases in myo-inositol levels in nerve and motor conduction velocity (Table I). Thus, it is concluded that the sodium-potassium adenosinetriphosphatase defect reflects some as yet unidentified disruption in inositol phospholipid metabolism in diabetic nerve [2]. Recent evidence suggests that metabolites of inositol phospholipid catabolism stimulate the activity of a cyclic adenosine monophosphate-independent, phospholipid-dependent and calcium-dependent protein kinase (protein kinase C) that activates sodium-potassium adenosinetriphosphatase, and that a deficiency in myo-inositol interferes with this sequence, producing a sodium-potassium adenosinetriphosphatase defect in diabetic nerve.
observed parallel responses of nerve myo-inositol content, sodium-potassium adenosinetriphosphatase activity, and motor nerve conduction velocity in animals with acute diabetes during insulin replacement or oral myo-inositol supplementation, in which parallelism persists despite continuing hyperglycemia and elevated concentrations of glucose, sorbitol, and fructose in nerve (Table I) [l ,10,16,17]. Thus, impairment of nerve sodium-potassium adenosinetriphosphatase activity and nerve conduction in animals with acute diabetes is a consequence of altered myo-inositol metabolism in the diabetic nerve, which secondarily alters the structure and/or function of the sodium-potassium adenosinetriphosphatase-membrane complex [18]. ,pimilar relationships between tissue sodium-potassium myo-inositol : metabolism and adenosinetriphosihatase activity have recently been reported for diabetic renal gibmerulus [2,3] and retina [4,5]. Relationship between Aldose Reductase Inhibitors and Myo-lnositol Metabolism. Aldose reductase inhibitors such as sorbinil completely prevent the decrease in myo-inositol content and altered sodium-potassium adenosinetriphosphatase activity in diabetic peripheral nerve and glomerulus [l-4]. As mentioned previously, the administration of myo-inositol to correct characteristically reduced tissue content without influencing the concentration of sorbitol in the nerve and glomerulus restores conduction velocity in diabetic nerve and increased urinary protein loss toward normal values [l-4,19]. Thus, increased polyol pathway activity in diabetes appears to contribute to the disruption in myo-inositol metabolism. Alternatively, aldose reductase inhibitors influence the metabolism of myo-inositol through a mechanism other than aldose reductase inhibition, and their action in diabetic tissue is attributable, at least in part, to their effects on the myo-inositol-related defect in sodium-potassium adenosinetriphosphatase discussed earlier. Hence, the two most commonly cited biochemical abnormalities in tissues susceptible to diabetic complications-altered myoinositol and sorbitol metabolism-are closely intertwined both metabolically and functionally.
November
ON ALDOSE
NERVE MYO-INOSITOL TRANSPORT AND SODIUMPOTASSIUM ADENOSINETRIPHOSPHATASE The peripheral nerve myo-inositol transport system is activated by the sodium gradient generated by sodium-potassium adenosinetriphosphatase [25]. Therefore, it might be expected that the reduced sodium-potassium adenosinetriphosphatase activity of diabetic peripheral nerve would secondarily impair the sodium-dependent uptake of myo-inositol. Recent experiments confirm that this is so and that sodium-dependent myo-inositol uptake remains
The
American
Journal
of Medicine
Volume
79 (suppl
5A)
15
SYMPOSIUM
ON ALDOSE REDUCTASE
INHIBITION-GREENE
and LAlTlMER
Hyperglycemia I
-4
Compititive inhibition
myo-inositol
fPolyol oathwav
uptake
P 4NalK ATPase activity
4Tissue myo-inositol &Membrane phosphoinositide
Figure 1. Proposed cyclic metabolic defect involving myoinositol, phosphoinositide, and the sodium-potassium adenosinetriphosphatase that is induced by hyperglycemia in peripheral nerve.
reduced in diabetic peripheral nerve in vitro even after tissue glucose concentrations are normalized. Furthermore, this impairment correlates with the reduction in sodiumpotassium adenosinetriphosphatase activity under various experimental conditions [3], suggesting that reduced sodium-potassium adenosinetriphosphatase activity in diabetic nerve further compromises endoneurial sodiumdependent uptake of myo-inositol. Thus, these relationships set the stage for a self-sustaining cyclic metabolic disruption involving myo-inositol and sodium-potassium adenosinetriphosphatase in diabetic peripheral nerve (Figure 1). A similar sequence may also exist in other tissues susceptible to diabetic complications. UNIFYING
METABOLIC
HYPOTHESIS
Hyperglycemia alters the metabolism in nerves in several ways. Competitive inhibition of sodium-dependent uptake of myo-inositol and/or increased polyol (sorbitol) pathway activity reduce the myo-inositol content of nerve, alter the phosphoinositide metabolism of nerve, diminish protein kihase C activity [26], and impair sodium-potassium adenosinetriphosphatase activity, further reducing sodium-gradient-dependent myo-inositol uptake and there-
by initiating a self-sustaining cycle. Impairment of the electrogenic sodium-potassium adenosinetriphosphatase activity acutely and secondarily reduces nerve conduction velocity [27] while probably also inhibiting other sodiumgradient-dependent processes such as amino a&d uptake and intracellular water homeostasis. Other defects may also occur as a result of alterations in the activity of inositol phospholipid metabolism or protein kinase C that are independent of sodium-potassium adenosinetriphosphatase and that may possibly involve other membranebound proteins, such as the voltage-dependent sodiumchannel or other membrane cationic adenosinetriphosphatases. These abnormalities may become self-sustaining cyclic defects, with potentially widespread pathophysiolbgic implications, which lead to impaired nerve conduction, axonal transport, intermediary-metabolism, and water balance, and possibly also contribute to the later structural defects in diabetic peripheral nerve. By analogy, a similar series of events may occur in other tissues in which the myo-inositol conteht is reduced through exposure to elevated concentrations of glucose. LINK BETWEEN P~LY~L METABOLISM AND THE MYO-IfiOSlTOL-RELATED SODIUkl-POTASSIUM ADENOSINETRIPHOSPHATASE DEFECT
Increased polyol pathway activity and the effects of aldose reductase inhibitors on peripheral nerve, renal glomerulus, retina, and arterial wall may be mediated primarily through the my&inositol-related sodium-potassium adenosinetriphosphatase defect described earlier. Elucidation of the link between sorbitol metabolism and myo-inositol or sodium-potassium adenosinetriphdsphatase is, therefoie, essential for comprehending the action -of aldose reductase inhibitors. Aldose reductase inhibition would be expected to alter the intracellular concentrations of several polyol pathway intermediates including sorbitol, fructose, oxidized nicotinamide-adenine dinucleotide phosphate, reduced nicotinamide-adenine dinucleotide phosphate, oxidized nicdtinamide-adenine dinucleotide, and reduced nicotinamide-adenine dinucleotide. Possible links between these intermediates and myo-inositol and sodium-potassium adenosinetriphosphatase metabolism remain to be elucidated.
REFERENCES 1.
2.
3.
16
Greene DA, Lattimer SA, Ulbrecht JS, Carroll PB: Glucose-induced alterations in nerve metabolism: current perspective on the pathogenesis of diabetic neuropathy and future directions for research and therapy. Diabetes Care 1985; 8: 290-299. Beyer-Mears A, Ku L, Cohen MP: Glomerular polyol accumulation in diabetes and its prevention by oral sorbinil. Diabetes 1984; 33: 604-607. Cohen MP: Reduced glomerular sodium-potassium
November 15, 1985
The American Journal of Medicine
4.
5. 6.
adenosinetriphosphatase activity in acute streptozotocin diabetes and its prevention by oral sorbinil. Diabetes (in press). MacGregor LC, Rosecan LR, Laties AM, Matchinsky FM: Microanalysis of total lipid, glucose, sorbitol, and myo-inositol in individual retinal layers of normal and alloxan diabetic rabbits (abstr). Diabetes 1984; 33(suppl 1): 89A. MacGregor L, Matchinsky F: Personal communication. Morrison AD: Aortic smooth muscle metabolism: effect of polyol
Volume 79 (suppl 5A)
SYMPOSIUM
7.
8.
9.
10.
11.
12. 13. 14.
15.
16.
17.
pathway inhibition (abstr). Clin Res 1984; 32(suppl 5): 851A. Mayhew JA, Gillon KRW, Hawthorne JN: Free and lipid inositol, sorbltol, and sugars in sciatic nerve obtained post mortem from diabetic and control subjects. Diabetologia 1983; 24: 1315. Fagius J, Jameson S: Effects of aldose reductase inhibitor treatment in diabetic polyneuropathy-a clinical and neurophysiological study. J Neurol Neurosurg Psychiatry 1981; 44: 9911001. Judzewitsch RG, Jaspan JB, Polonsky KS, et al: Aldose reductase inhibition improves nerve conduction velocity in diabetic patients. N Engl J Med 1983; 308: 119-125. Greene DA, DeJesus PV, Winegrad Al: Effects of insulin and dietary myo-inositol on impaired peripheral motor nerve conduction velocity in, acute streptozotocin diabetes. J Clin Invest 1975; 55: 1326-1336. Jakobsen J: Early and preventable changes of peripheral nerve structure and function in insulin-deficient diabetic rats. J Neural Neurosurg Psychiatry 1979; 42: 502-518. Sima AAF: Peripheral neuropathy in the spontaneously diabetic BB-Wistar rat. Acta Neuropathol (Ben) 1980; 51: 223-227. Sharma AK, Thomas PK: Peripheral nerve structure and function in experimental diabetes. J Neurol Sci 1974; 23: l-15. Brown MJ, Sumner AJ, Greene DA, et al: Distal neuropathy in experimental diabetes mellitus. Ann Neurol 1980; 8: 168178. Greene DA: Metabolic abnormalities in diabetic peripheral nerve: relation to impaired function. Metabolism 1983; 32 (suppl 1): 118-l 23. Greene DA, Lewis RA, Lattimer SA, Brown MJ: Selective effects of myo-inositol administration on sciatic and tibia1 motor nerve conduction parameters in the streptozotocin-diabetic rat. Diabetes 1982; 31: 573-578. Mayer JH, Tomlinson DR: Prevention of defects of axonal trans-
November
15, 1985
18.
19.
20.
21,
22.
23.
24.
25.
26.
27.
The
ON ALDOSE
REDUCTASE
INHIBITION-GREENE
and LATTIMER
port and nerve conduction velocity by oral administration of myo-inositol or an aldose reductase inhibitor in streptozotocin-diabetic rats. Diabetologia 1983; 25: 433-438. Greene DA, Yagihashi S, Lattimer SA, Sima AAF: Nerve Na+K+ATPase, conduction, and myo-inositol in the insulindeficient BB rat. Am J Physiol 1984; 247: E534-E539. Beyer-Mears A, Ku L, Cohen MP: Glomerular polyol accumulation in diabetes and its prevention with sorbinil (abstr). Diabetes 1984; 33 (suppl 1): 89A. Trachtenberg MC, Packey DJ, Sweeney T: In vivo functioning of the NA+K+-activated ATPase. Curr Top Cell Regul 1981; 19: 159-217. Lingham RB, Sen AK: Regulation of rat brain (Na+K+)-ATPase activity by cyclic AMP. Biochim Biophys Acta 1982; 688: 475485. Greene DA, Winegrad Al: Effects of acute experimental diabetes on composite energy metabolism in peripheral nerve axons and Schwann cells. Diabetes 1981; 30: 967-974. Das PK, Bray G, Aguayo AJ, Rasminsky M: Diminished ouabain-sensitive sodium-potassium ATPase activity in sciatic nerves of rats with streptozotocin-induced diabetes. Exp Neural 1976; 53: 285-288. Palmano KP, Whiting PH, Hawthorne JN: Free and lipid myoinositol in tissues from rats with acute and less severe streptozotocin-induced diabetes. Biochem J 1977; 167: 229-235. Greene DA, Lattimer SA: Sodium and energy dependent uptake of myo-inositol by rabbit peripheral nerve. J Clin Invest 1972; 70: 1009-1018. Greene DA, Lattimer SA: Phorbol ester corrects the myo-inositol-related Na/K-ATPase defect in diabetic nerve (abstr). Clin Res 1985; 33: 569A. Brismar T, Sima AAF: Changes in nodal function in nerve fibres of the spontaneously diabetic BB-Wistar rat: potential clamp analysis. Acta Physiol Stand 1981; 113: 499-506.
American
Journal
of Medicine
Volume
79 (suppl
5A)
17