- Email: [email protected]
Elevated l -xylulose concentrations in serum: A difference between type I and type II diabetes
Elevated L-Xylulose A Difference Between Thomas
J. Merimee,
Concentrations in Serum: Type I and Type II Diabetes Robert
I. Misbin,
and
Larry
Gold
L-Xylulose, which can be derived from glucose directly or from mucopolysaccharide degradation, was measured in serum samples from 61 diabetics and 42 controls. All serum samples from the controls were negative for L-xylulose. Fifteen of 30 adult-onset diabetics, in contrast to only four of 31 juvenile-onset diabetics, had detectable L-xylulose levels. This difference between adult-onset and juvenile-onset diabetics was significant at the 0.001 level. Detectability of L-xylulose in serum did not appear to be influenced by the fasting mean or peak serum concentration of glucose. Mean serum concentrations of growth hormone did not correlate with L-xylulose levels. The reason for the variation of L-xylulose between type-l and type-11diabetic subjects could not be identified. We would postulate a variation in the degradation of glycosaminoglycan. These results support the view that type-l and type-11diabetes are different diseases.
T
HE
NONPHOSPHORYLATED SUGAR, Lis an intermediate of the glucuronic acid cycle, derived either from glucose directly or from metabolites of glycoprotein degradation. Our interest in this compound was stimulated over a decade ago by reports from Winegrad and Burden’ that fasting serum concentrations of L-xylulose were higher in diabetics than in normal subjects. The major finding of the present study is that a significant difference exists in serum L-xylulose levels between type-1 and type-II diabetics. This difference may relate to a diminished ability of type-1 diabetics to degrade acid mucopolysaccharides. xylulose,
MATERIALS
AND
METHODS
Sixty-one diabetics and 42 healthy controls were admitted to the Clinical Research Center of the University of Florida after proper informed consent was obtained. All of the diabetics were clinically stable, and none had been acutely ill within 3 months of the study period. Thirty-one of the subjects had type-I, or juvenile-onset diabetes. This diagnosis was based on (I) onset of diabetes before the age of IO years in 15 subjects, and before the age of 20 years in eight additional patients. Eight diabetics in whom the disease appeared between ages 30 and 40 years were considered type I in onset on the basis of onset with ketoacidosis, demonstrable episodes of acidosis following the initial diagnosis, and a reliable history indicating an absolute need for exogenous insulin that began and persisted after
From the Division of Endocrinology and Metabolism, Department of Medicine. University of Florida School of Medicine, Gainesville, Florida. Received for publication March 28, 1983. This work was supported in part by a grant from the National Institutes of Health. Division of Arthritis, Metabolism and Digestive Diseases (AM-18130). The authors wish to acknowledge support by a grant from the General Clinical Research Centers Program of the Divison of Research Resources, National Institutes of Health (RR-00082). and gifts from the Upjohn Company and Theodore C. Presser. Dr Misbin is the recipient of a Special Emphasis Research Career Award (1 KOI AM 00.561-01). Address reprint requests to Dr Thomas J. Merimee. Division of Endocrinology and Metabolism, Department of Medicine, University of Florida School of Medicine, Gainesville, FL 32610. 0 1984 by Grune & Stratton, Inc. 0026-0495/84/3301~013$01.00/0 a2
the initial discovery of the diabetic state. Thirty diabetics with type-II (adult-onset) diabetes were identified. Subjects placed in this category met the following qualifications: (I) they were not hospitalized for ketoacidosis in the first 3 years of their disease, and (2) they required no insulin for management for periods varying from 3 to I7 years after the initial diagnosis of diabetes. At the time of the study, however, 22 of the latter subjects were receiving exogenous insulin treatment. The diabetics and controls were maintained ambulatory during this 24-hour admission, and diets comparable to those normally ingested were continued. A soft catheter was placed in a wrist vein, and samples were collected via a heparin lock. Serum samples were obtained every two hours for 24 hours. A 2-mL serum aliquot was immediately deproteinized with 0.2 mL of 70% perchloric acid, frozen, and neutralized with potassium hydroxide just before assay for L-xylulose. A second serum aliquot was utilized for the measurement of glucose and immunoreactive growth hormone. These were determined by methods previously described in detail.2a The method used to measure L-xylulose was based on the enzymatic assay described by Touster et al5 and Hickman and Ashwell. The incubation mixture contained 0.4 mL TRIS buffer, 0.1 mol/L, pH 7.0; 0.01 mL fresh NADPH, type IO (5 mg/mL); 0.020 mL of fresh cysteine HCI (50 mmol/L); 0.100 mL of 50 mmol/L MgCI,; and 0.45 mL of sample. The reaction was started by the addition of 0.02 mL of L-xylulose reductase (0.2 U/mL, Sigma Chemical Co, St Louis, MO). The enzyme was a commercial pigeon liver preparation that differed from that employed in other laboratories. The disappearance of NADPH was measured at 340 nm at three and six minutes. An L-xylulose standard (generously supplied by G. Ashwell) was assayed with each set of samples to ensure that the reaction had gone to completion (90%). The reaction had usually gone to completion by three minutes. Allowing the reaction to go longer decreased the reliability of the assay because of drift in the blank. Because of the very small quantity of L-xylulose in plasma and spontaneous oxidation of NADPH, we found it necessary to perform a blank containing deproteinized plasma but no enzyme for each sample. Samples from diabetic patients and normal controls were always assayed at the same time to eliminate any possible bias; ie, each sample was assayed on three different occasions. A sample was said to be “positive” for L-xylulose only if L-xylulose was detected in three separate assay runs. This criterion was based on preliminary data, which showed that when L-xylulose was added to sera, it was always detected in all three assays when the concentration exceeded 0.4 mg/dL. At lower concentrations the detection rate was variable. RESULTS L-Xylulose
Eleven of 30 adult-onset ing
L-xylulose
levels
and
diabetics had positive fastof 31 juvenile-onset
three
Metabo/ism,
Vol.
33, No.
1
(Januaryl,
1984
83
L-XYLULOSE IN DIABETES
diabetics had positive samples, but none of the 42 controls had detectable L-xylulose levels in the fasting state. In the subjects in whom L-xylulose was detectable, there was a wide range of values. L-Xylulose concentrations ranged from 0.3 to 0.7 mg/dL in type-1 diabetic patients and from 0.1 to 0.8 mg/dL in type-II diabetic patients. Comparisons of the means of the data were not possible because of the small number of type-1 diabetics (n = 3) with detectable L-xylulose levels. Chi-square analysis of these data, however, yielded a P value of 0.01 when comparing fasting L-xylulose in type-II versus type-1 diabetics. When all 13 serum specimens per subject obtained in the 24-hour sampling period were assayed for Lxylulose, the difference between type-1 onset and typeII diabetes was even more significant. Fifteen type-II and four type-1 subjects had one to four samples positive per 24-hour period. The percentage of each group with detectable L-xylulose levels at some time during the day is shown in Fig. 1. This difference between subjects with type-II diabetes and type-1 diabetes was significant at the 0.001 level. (Chi-square analysis: x2 = 7.61, df= 1, P = 0.001). Glucose
and Growth
Hormone
Fasting and mean serum concentrations of glucose over 24 hours were determined from a complete set of
ADULT ONSET DIABETICS
JUVENILE DIABETICS
CONTROLS
JODM 0465 0423
1
. i?
B .
. L-XYLULOSE
POSITIVE
0 L-XYLULOSE
NEGATIVE
Fig. 2. The relationship of the mean serum glucose concentration to the L-xylulose concentration in subjects “positive” for L-xylulose is shown. No significant correlation was noted.
samples (13) on 30 type-II and 20 type-1 diabetics. The mean serum glucose concentration was not related to L-xylulose detectability in either type-II or type-1 patients. The mean & SEM of the glucose levels for all L-xylulose-positive and all L-xylulose-negative patients were 184 + 17 mg/dL and 235 + 14 mg/dL, respectively (P > 0.10). There was no correlation between fasting or mean serum glucose and fasting L-xylulose in the basal state. Data for the mean glucose concentrations are illustrated in Fig. 2. Mean and fasting growth hormone concentrations in serum were lower in the adult-onset diabetics because of the higher percentage of obese individuals in this group. Neither increases nor decreases of serum growth hormone concentrations could be correlated with times at which samples were positive for Lxylulose. DISCUSSION
Fig. 1. The percent of subjects in each group with detectable L-xylulose concentrations in serum is shown. Serum samples collected every two hours ware assayed for L-xylulosa. The difference in detectability of L-xylulose between juvenile-onset and adult-onset diabetics was significant with a P value of 0.001.
L-Xylulose can be derived in vivo from three principal sources (Fig. 3): (1) by de novo synthesis from glucose or UDG-glucose via the glucuronic acid cycle;’ (2) from the enzymatic cleavage of inositol to yield glucuronate, also an intermediate of the glucuronic acid cycle;8 and, (3) from the metabolism of products derived from mucopolysaccharide degradation-primarily uranic acid.’ Each process requires the glucuranic acid cycle for the actual formation of Lxylulose. Although it was initially conceivable to us that the first reaction above might explain our findings, we now believe this unlikely. We find, for example, no relationship between severity of hyperglycemia and detectability of L-xylulose, and there is no relationship of L-
84
MERIMEE, MISBIN, AND GOLD
Fig. 3. The glucuronic acid cycle is shown to illustrate the principal entry points. Multiple intermediary steps are not shown for the purpose of clarity bee text I.
xylulose to meal ingestion. A similar observation was made on the secretion of growth hormone, ie, Lxylulose did not correlate with either the peak or mean serum concentrations of growth hormone. The enzymatic cleavage of inositol with conversion to D-glucuronic acid, a precursor of L-xylulose in the glucuronic acid pathway is likewise theoretically possible. In other studies, however, oral loads of inositol administered to both normal and diabetic subjects failed to cause a rise in either serum or urinary L-xylulose concentrations (Clements R, Winegrad A, personal communication, December, 1977). An additional possibility, and one we favor, is that the L-xylulose measured in the diabetic is derived from mucopolysaccharides. Acid mucopolysaccharides (glycosaminoglycans) are covalently bound to protein, with each glycosaminoglycan chain containing a large number of identical, repeating, disaccharide units, the
most common disaccharide unit being D-glucuronic acid. Mammalian tissues contain group-specific enzymes that catalyze the hydrolysis of the B-glucuronides to yield various aglycans and free glucuronic acid, and the latter, as indicated in Fig. 3, is readily converted into L-xylulose. In the normal individual, the rates of synthesis and degradation of mucopolysaccharides is such that little L-xylulose is formed. Our data would thus imply some difference between juvenile-onset and adult-onset diabetics in the capacity to either degrade glycosaminoglycans or to clear them from the body. There is some evidence, however, that the liver may be a major source of L-xylulose, and we have no firm basis for excluding the possibility that altered hepatic function is involved. The observation that abnormally high values appear to be more prevalent in adult-onset diabetics than in type-1 diabetics reinforces the current view that these diseases have a different genetic basis.” More definitive studies of this problem are required, particularly comparative studies of glycosaminoglycan synthesis and degradation in particular tissues in juvenile-onset and adult-onset diabetics. Although accumulation of glycosaminoglycans is recorded in tissues of diabetics, kinetic studies of this problem are sadly lacking.“-14 It is possible that the deposition of glycosaminoglycans in tissues such as the myocardium is a problem of degradation rather than one partially or solely of abnormal synthesis. The elevated levels of L-xylulose in type-II diabetics versus type-1 diabetics supports the concept that has emerged from genetic studies that type-1 and type-II diabetics are different diseases.
REFERENCES 1. Winegrad AI, Burden C: L-xylulose metabolism in diabetes mellitus. N Engl J Med 274:298-305, 1966 2. Hugges AG, Nixon DA: Use of glucose oxidase, peroxidase and odianesidine in determination of blood and urinary glucose. Lancet 2:368, 1957 3. Click SM, Roth J, Yallow RS, et al: Immunoassay of human growth hormones in plasma. Nature 199:784-787, 1963 4. Herbert V, Lau K, Gottlieb CW, et al: Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375-1380, 1965 5. Touster 0, Reynolds VH, Hutcheson RM: The reduction of L-xylulose to xylitol by guinea pig liver mitochondria. J Biol Chem 221:697-709,1956 6. Hickman J, Ashwell G: A sensitive and stereospecific enzymatic assay for xylulose. J Biol Chem 234:758-761, 1959 7. Hollmann S, Touster 0: An enzymatic pathway from Lxylulose to p-xylulose. J Am Chem Sot 78:35443545, 1956 8. Lewy GA, Conchic J: B-glucuronidase and the hydrolysis of glucuronides, in Dutton GH (ed): Glucuronic Acid: Free and Combined. London, Academic Press, 1966, pp 301-364
9. Burns JJ, Conney AH: Metabolism of glucuronic acid and its lactone, in Dutton GH (ed): Glucuronic Acid: Free and Combined. London, Academic Press, 1966, pp 365-384 10. Cahill GF, McDevitt HO: Insulin-dependent tus: The initial lesion. N Engl J Med 304:14541465,
diabetes 1981
melli-
11. Haider B, Yeh CK, Thomas G, et al: Altered myocardial function and collagen in diabetic rhesus monkeys on atheaogenic diet. Clin Res 26:547A, 1978 12. Westberg NG: Biochemical alterations of the human glomerular basement membrane in diabetes. Diabetes (suppl 2):92&925, 1976 13. Beisinger PJ, Spiro RG: Studies on the human glomerular basement membrane: Composition, nature of the carbohydrate units and chemical changes in diabetes mellitus. Diabetes 22:18&183, 1973 14. Spiro RG, Spiro MG: Effect of diabetes on the biosynthesis of the renal glomerular basement membranes: Studies of the glucosyltransferase. Diabetes 20:641648, 1971
J. Merimee,
Concentrations in Serum: Type I and Type II Diabetes Robert
I. Misbin,
and
Larry
Gold
L-Xylulose, which can be derived from glucose directly or from mucopolysaccharide degradation, was measured in serum samples from 61 diabetics and 42 controls. All serum samples from the controls were negative for L-xylulose. Fifteen of 30 adult-onset diabetics, in contrast to only four of 31 juvenile-onset diabetics, had detectable L-xylulose levels. This difference between adult-onset and juvenile-onset diabetics was significant at the 0.001 level. Detectability of L-xylulose in serum did not appear to be influenced by the fasting mean or peak serum concentration of glucose. Mean serum concentrations of growth hormone did not correlate with L-xylulose levels. The reason for the variation of L-xylulose between type-l and type-11diabetic subjects could not be identified. We would postulate a variation in the degradation of glycosaminoglycan. These results support the view that type-l and type-11diabetes are different diseases.
T
HE
NONPHOSPHORYLATED SUGAR, Lis an intermediate of the glucuronic acid cycle, derived either from glucose directly or from metabolites of glycoprotein degradation. Our interest in this compound was stimulated over a decade ago by reports from Winegrad and Burden’ that fasting serum concentrations of L-xylulose were higher in diabetics than in normal subjects. The major finding of the present study is that a significant difference exists in serum L-xylulose levels between type-1 and type-II diabetics. This difference may relate to a diminished ability of type-1 diabetics to degrade acid mucopolysaccharides. xylulose,
MATERIALS
AND
METHODS
Sixty-one diabetics and 42 healthy controls were admitted to the Clinical Research Center of the University of Florida after proper informed consent was obtained. All of the diabetics were clinically stable, and none had been acutely ill within 3 months of the study period. Thirty-one of the subjects had type-I, or juvenile-onset diabetes. This diagnosis was based on (I) onset of diabetes before the age of IO years in 15 subjects, and before the age of 20 years in eight additional patients. Eight diabetics in whom the disease appeared between ages 30 and 40 years were considered type I in onset on the basis of onset with ketoacidosis, demonstrable episodes of acidosis following the initial diagnosis, and a reliable history indicating an absolute need for exogenous insulin that began and persisted after
From the Division of Endocrinology and Metabolism, Department of Medicine. University of Florida School of Medicine, Gainesville, Florida. Received for publication March 28, 1983. This work was supported in part by a grant from the National Institutes of Health. Division of Arthritis, Metabolism and Digestive Diseases (AM-18130). The authors wish to acknowledge support by a grant from the General Clinical Research Centers Program of the Divison of Research Resources, National Institutes of Health (RR-00082). and gifts from the Upjohn Company and Theodore C. Presser. Dr Misbin is the recipient of a Special Emphasis Research Career Award (1 KOI AM 00.561-01). Address reprint requests to Dr Thomas J. Merimee. Division of Endocrinology and Metabolism, Department of Medicine, University of Florida School of Medicine, Gainesville, FL 32610. 0 1984 by Grune & Stratton, Inc. 0026-0495/84/3301~013$01.00/0 a2
the initial discovery of the diabetic state. Thirty diabetics with type-II (adult-onset) diabetes were identified. Subjects placed in this category met the following qualifications: (I) they were not hospitalized for ketoacidosis in the first 3 years of their disease, and (2) they required no insulin for management for periods varying from 3 to I7 years after the initial diagnosis of diabetes. At the time of the study, however, 22 of the latter subjects were receiving exogenous insulin treatment. The diabetics and controls were maintained ambulatory during this 24-hour admission, and diets comparable to those normally ingested were continued. A soft catheter was placed in a wrist vein, and samples were collected via a heparin lock. Serum samples were obtained every two hours for 24 hours. A 2-mL serum aliquot was immediately deproteinized with 0.2 mL of 70% perchloric acid, frozen, and neutralized with potassium hydroxide just before assay for L-xylulose. A second serum aliquot was utilized for the measurement of glucose and immunoreactive growth hormone. These were determined by methods previously described in detail.2a The method used to measure L-xylulose was based on the enzymatic assay described by Touster et al5 and Hickman and Ashwell. The incubation mixture contained 0.4 mL TRIS buffer, 0.1 mol/L, pH 7.0; 0.01 mL fresh NADPH, type IO (5 mg/mL); 0.020 mL of fresh cysteine HCI (50 mmol/L); 0.100 mL of 50 mmol/L MgCI,; and 0.45 mL of sample. The reaction was started by the addition of 0.02 mL of L-xylulose reductase (0.2 U/mL, Sigma Chemical Co, St Louis, MO). The enzyme was a commercial pigeon liver preparation that differed from that employed in other laboratories. The disappearance of NADPH was measured at 340 nm at three and six minutes. An L-xylulose standard (generously supplied by G. Ashwell) was assayed with each set of samples to ensure that the reaction had gone to completion (90%). The reaction had usually gone to completion by three minutes. Allowing the reaction to go longer decreased the reliability of the assay because of drift in the blank. Because of the very small quantity of L-xylulose in plasma and spontaneous oxidation of NADPH, we found it necessary to perform a blank containing deproteinized plasma but no enzyme for each sample. Samples from diabetic patients and normal controls were always assayed at the same time to eliminate any possible bias; ie, each sample was assayed on three different occasions. A sample was said to be “positive” for L-xylulose only if L-xylulose was detected in three separate assay runs. This criterion was based on preliminary data, which showed that when L-xylulose was added to sera, it was always detected in all three assays when the concentration exceeded 0.4 mg/dL. At lower concentrations the detection rate was variable. RESULTS L-Xylulose
Eleven of 30 adult-onset ing
L-xylulose
levels
and
diabetics had positive fastof 31 juvenile-onset
three
Metabo/ism,
Vol.
33, No.
1
(Januaryl,
1984
83
L-XYLULOSE IN DIABETES
diabetics had positive samples, but none of the 42 controls had detectable L-xylulose levels in the fasting state. In the subjects in whom L-xylulose was detectable, there was a wide range of values. L-Xylulose concentrations ranged from 0.3 to 0.7 mg/dL in type-1 diabetic patients and from 0.1 to 0.8 mg/dL in type-II diabetic patients. Comparisons of the means of the data were not possible because of the small number of type-1 diabetics (n = 3) with detectable L-xylulose levels. Chi-square analysis of these data, however, yielded a P value of 0.01 when comparing fasting L-xylulose in type-II versus type-1 diabetics. When all 13 serum specimens per subject obtained in the 24-hour sampling period were assayed for Lxylulose, the difference between type-1 onset and typeII diabetes was even more significant. Fifteen type-II and four type-1 subjects had one to four samples positive per 24-hour period. The percentage of each group with detectable L-xylulose levels at some time during the day is shown in Fig. 1. This difference between subjects with type-II diabetes and type-1 diabetes was significant at the 0.001 level. (Chi-square analysis: x2 = 7.61, df= 1, P = 0.001). Glucose
and Growth
Hormone
Fasting and mean serum concentrations of glucose over 24 hours were determined from a complete set of
ADULT ONSET DIABETICS
JUVENILE DIABETICS
CONTROLS
JODM 0465 0423
1
. i?
B .
. L-XYLULOSE
POSITIVE
0 L-XYLULOSE
NEGATIVE
Fig. 2. The relationship of the mean serum glucose concentration to the L-xylulose concentration in subjects “positive” for L-xylulose is shown. No significant correlation was noted.
samples (13) on 30 type-II and 20 type-1 diabetics. The mean serum glucose concentration was not related to L-xylulose detectability in either type-II or type-1 patients. The mean & SEM of the glucose levels for all L-xylulose-positive and all L-xylulose-negative patients were 184 + 17 mg/dL and 235 + 14 mg/dL, respectively (P > 0.10). There was no correlation between fasting or mean serum glucose and fasting L-xylulose in the basal state. Data for the mean glucose concentrations are illustrated in Fig. 2. Mean and fasting growth hormone concentrations in serum were lower in the adult-onset diabetics because of the higher percentage of obese individuals in this group. Neither increases nor decreases of serum growth hormone concentrations could be correlated with times at which samples were positive for Lxylulose. DISCUSSION
Fig. 1. The percent of subjects in each group with detectable L-xylulose concentrations in serum is shown. Serum samples collected every two hours ware assayed for L-xylulosa. The difference in detectability of L-xylulose between juvenile-onset and adult-onset diabetics was significant with a P value of 0.001.
L-Xylulose can be derived in vivo from three principal sources (Fig. 3): (1) by de novo synthesis from glucose or UDG-glucose via the glucuronic acid cycle;’ (2) from the enzymatic cleavage of inositol to yield glucuronate, also an intermediate of the glucuronic acid cycle;8 and, (3) from the metabolism of products derived from mucopolysaccharide degradation-primarily uranic acid.’ Each process requires the glucuranic acid cycle for the actual formation of Lxylulose. Although it was initially conceivable to us that the first reaction above might explain our findings, we now believe this unlikely. We find, for example, no relationship between severity of hyperglycemia and detectability of L-xylulose, and there is no relationship of L-
84
MERIMEE, MISBIN, AND GOLD
Fig. 3. The glucuronic acid cycle is shown to illustrate the principal entry points. Multiple intermediary steps are not shown for the purpose of clarity bee text I.
xylulose to meal ingestion. A similar observation was made on the secretion of growth hormone, ie, Lxylulose did not correlate with either the peak or mean serum concentrations of growth hormone. The enzymatic cleavage of inositol with conversion to D-glucuronic acid, a precursor of L-xylulose in the glucuronic acid pathway is likewise theoretically possible. In other studies, however, oral loads of inositol administered to both normal and diabetic subjects failed to cause a rise in either serum or urinary L-xylulose concentrations (Clements R, Winegrad A, personal communication, December, 1977). An additional possibility, and one we favor, is that the L-xylulose measured in the diabetic is derived from mucopolysaccharides. Acid mucopolysaccharides (glycosaminoglycans) are covalently bound to protein, with each glycosaminoglycan chain containing a large number of identical, repeating, disaccharide units, the
most common disaccharide unit being D-glucuronic acid. Mammalian tissues contain group-specific enzymes that catalyze the hydrolysis of the B-glucuronides to yield various aglycans and free glucuronic acid, and the latter, as indicated in Fig. 3, is readily converted into L-xylulose. In the normal individual, the rates of synthesis and degradation of mucopolysaccharides is such that little L-xylulose is formed. Our data would thus imply some difference between juvenile-onset and adult-onset diabetics in the capacity to either degrade glycosaminoglycans or to clear them from the body. There is some evidence, however, that the liver may be a major source of L-xylulose, and we have no firm basis for excluding the possibility that altered hepatic function is involved. The observation that abnormally high values appear to be more prevalent in adult-onset diabetics than in type-1 diabetics reinforces the current view that these diseases have a different genetic basis.” More definitive studies of this problem are required, particularly comparative studies of glycosaminoglycan synthesis and degradation in particular tissues in juvenile-onset and adult-onset diabetics. Although accumulation of glycosaminoglycans is recorded in tissues of diabetics, kinetic studies of this problem are sadly lacking.“-14 It is possible that the deposition of glycosaminoglycans in tissues such as the myocardium is a problem of degradation rather than one partially or solely of abnormal synthesis. The elevated levels of L-xylulose in type-II diabetics versus type-1 diabetics supports the concept that has emerged from genetic studies that type-1 and type-II diabetics are different diseases.
REFERENCES 1. Winegrad AI, Burden C: L-xylulose metabolism in diabetes mellitus. N Engl J Med 274:298-305, 1966 2. Hugges AG, Nixon DA: Use of glucose oxidase, peroxidase and odianesidine in determination of blood and urinary glucose. Lancet 2:368, 1957 3. Click SM, Roth J, Yallow RS, et al: Immunoassay of human growth hormones in plasma. Nature 199:784-787, 1963 4. Herbert V, Lau K, Gottlieb CW, et al: Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375-1380, 1965 5. Touster 0, Reynolds VH, Hutcheson RM: The reduction of L-xylulose to xylitol by guinea pig liver mitochondria. J Biol Chem 221:697-709,1956 6. Hickman J, Ashwell G: A sensitive and stereospecific enzymatic assay for xylulose. J Biol Chem 234:758-761, 1959 7. Hollmann S, Touster 0: An enzymatic pathway from Lxylulose to p-xylulose. J Am Chem Sot 78:35443545, 1956 8. Lewy GA, Conchic J: B-glucuronidase and the hydrolysis of glucuronides, in Dutton GH (ed): Glucuronic Acid: Free and Combined. London, Academic Press, 1966, pp 301-364
9. Burns JJ, Conney AH: Metabolism of glucuronic acid and its lactone, in Dutton GH (ed): Glucuronic Acid: Free and Combined. London, Academic Press, 1966, pp 365-384 10. Cahill GF, McDevitt HO: Insulin-dependent tus: The initial lesion. N Engl J Med 304:14541465,
diabetes 1981
melli-
11. Haider B, Yeh CK, Thomas G, et al: Altered myocardial function and collagen in diabetic rhesus monkeys on atheaogenic diet. Clin Res 26:547A, 1978 12. Westberg NG: Biochemical alterations of the human glomerular basement membrane in diabetes. Diabetes (suppl 2):92&925, 1976 13. Beisinger PJ, Spiro RG: Studies on the human glomerular basement membrane: Composition, nature of the carbohydrate units and chemical changes in diabetes mellitus. Diabetes 22:18&183, 1973 14. Spiro RG, Spiro MG: Effect of diabetes on the biosynthesis of the renal glomerular basement membranes: Studies of the glucosyltransferase. Diabetes 20:641648, 1971