Yasuhiro Tahara

Kinetics of HbA1c, Glycated Albumin, and Fructosamine and Analysis of Their Weight Functions Against Preceding Plasma Glucose Level

OBJECTIVE To examine the kinetics of HbA1c, glycated albumin (GA), and fructosamine (FA) levels in response to plasma glucose change and their relationship with the preceding plasma glucose level. RESEARCH DESIGN AND METHODS The time courses of HbA1c, GA, and FA after acute glycemic normalization were observed in nine patients with newly diagnosed non-insulin-dependent diabetes mellitus and compared with theoretical ones. Their weight functions against preceding plasma glucose level were analyzed assuming a stepwise plasma glucose change and compared with the theoretical prediction. RESULTS The fasting plasma glucose level was acutely normalized after admission with a half-time of 6.3 ± 2.4 days (mean ± SD). The HbA1c level decreased linearly during the initial 2 months with a half-time of 34.6 ± 10.1 days, followed by a gradual decrease thereafter. GA and FA levels decreased very rapidly during the initial 2–3 weeks with half-times of 17.1 ± 2.8 and 12.2 ± 4.8 days, respectively, followed by a gradual decrease thereafter. The time courses of HbA1c, GA, and FA agreed well with theoretically estimated decay curves. Experimental values of weight functions against the preceding plasma glucose level agreed well with the theoretical prediction. The weight functions for glycated proteins had maximum values on the days just before the measurement of glycated proteins and gradually decreased with an increasing time interval. The lengths of the periods over which the weight functions for HbA1c, GA, and FA extend back were estimated to be roughly 100, 40, and 30 days, respectively. CONCLUSIONS The levels of HbA1c, GA, and FA do not reflect the simple mean but reflect the weighted mean of the preceding plasma glucose level over a considerably longer period than was previously speculated.

Correlation Between Minimal Secretory Capacity of Pancreatic β-Cells and Stability of Diabetic Control

The significance of the minimal secretory capacity of pancreatic β-cells for the stability of the plasma glucose level was studied in 20 patients with insulin-dependent diabetes mellitus. Changes in plasma concentrations of major counterregulatory hormones in response to hypoglycemia were also investigated in these patients to clarify their contribution to diabetic brittleness. β-Cell function was evaluated on the basis of elevation of plasma C-peptide immunoreactivity (CPR) during the intravenous glucagon test with a highly sensitive assay for plasma CPR that could detect as little as 0.03 ng/ml. After stimulation with glucagon, a significant increase in plasma CPR was observed in 10 of the patients whose β-cell function had been evaluated as completely depleted by a conventional assay for plasma CPR. A clear inverse correlation was found between the secretory capacity of pancreatic β-cells measured in this way and the degree of glycemic instability (r = −.74, P < .01). Infusion of insulin at a rate of 0.15 U.kg−1.h−1 for 60 min caused a continuous decrease in the plasma glucose level, resulting in neuroglycopenia in 7 of the 10 CPR nonresponders but only 2 of the CPR responders. During insulin-induced hypoglycemia, plasma glucagon immunoreactivity did not increase in the CPR nonresponders but increased significantly in the CPR responders. A positive correlation was found between the minimal residual β-cell capacity and the responsiveness of α-cells to hypoglycemia (r = .65, P < .01).In contrast to the difference in the responses of their pancreatic α-cells to hypoglycemia, the two groups showed more or less the same responses of plasma epinephrine, norepinephrine, growth hormone, and cortisol to hypoglycemia. Total lack of insulinogenic reserve inevitably results in loss of automatic regulation of the circulating insulin level and seems to be a major factor in causing hyperlability of diabetic control. The lack of β-cell function may be related causally to pancreatic α-cell dysfunction, which also contributes in part to metabolic variability in brittle diabetes.

The Response of GHb to Stepwise Plasma Glucose Change Over Time in Diabetic Patients

G lycation of red cell hemoglobin and serum proteins is increased in diabetic patients by chronic hyperglycemia. Because glycation takes place throughout the life span of these proteins, the levels of glycated proteins provide information about the degree of hyperglycemia, during their life span. GHb is generally considered to reflect the mean plasma glucose level in the preceding 2 3 mo, whereas glycated albumin is considered to reflect the mean plasma glucose level in the preceding 2 -3 wk. However, this idea is not clearly verified because few reports answered the question of exactly how the plasma glucose level, in the preceding period, contributes to the level of glycated proteins. In a recent paper (1), we analyzed the relationship between the level of glycated protein and the preceding plasma glucose level using a linear kinetic model. We derived a simple, generalized formula, which describes how the plasma glucose level in the preceding period contributes to the level of glycated protein. The results indicate that the level of glycated protein reflects the weighted mean plasma glucose level found in the past. To examine whether or not this result is applicable to GHb levels in humans, we analyzed responses of HbAlc levels to stepwise plasma glucose decrements in 10 NIDDM patients. Glycemic control of these patients was very poor on admission: FPG > 11 mM and HbAlc > 10%. Their plasma glucose level was rapidly normalized after admission, and FPG was kept <8 mM over 4 mo. Four patients were treated with insulin, 5 with oral hypoglycemic agent, and 1 with diet alone. Plasma glucose was measured by the glucose oxidase method. HbAlc was measured by the HPLC method using HLC-723GHb (Tosoh, Tokyo, Japan). The mean ± SD FPG level on admission was 15.5 ± 2.6 mM and was rapidly reduced to 10.2 ± 2.6, 8.1 ± 2.0, 7.0 ± 1.2, 6.6 ± 0.7, 7.0 ± 1.0, 6.6 ± 0.8, and 6.4 ± 0.7 mM after 1, 2, 3, 4, 8, 12, and 16 wk, respectively. The half-time of FPG change was 6.9 ± 2.6 days. The HbAlc level was 12.7 ± 2.5% on admission and was gradually reduced to 9.9 ± 1.7, 8.3 ± 1 . 1 , 7.3 ± 0.8, and 6.6 ± 0.6% after 4, 8, 12, and 16 wk, respectively. The proportion of the total change in HbAlc during 4, 8, 12, and 16 wk was 45 ± 7, 71 ± 7, 87 ± 6, and 98 ± 4% (theoretically 41, 72, 91, and 99.6%, respectively). The times for 50 and 75% HbAlc change were 32 ± 6 and 63 ± 11 days (theoretically, 35 and 60 days). The time course of HbAlc change closely agreed with the theoretical prediction (Fig. 1).

Analysis of Early-Phase Insulin Responses in Nonobese Subjects With Mild Glucose Intolerance

OBJECTIVE To study the possible contribution of a B-cell defect in the development of glucose intolerance in nonobese subjects. RESEARCH DESIGN AND METHODS There were 41 normal, nondiabetic subjects; 18 subjects with IGT; and 21 patients with NIDDM. All subjects were nonobese (BMI < 27 kg/m2). Insulin secretory responses to an OGTT, IVGTT, and GST were studied. RESULTS Early-phase insulin responses to OGTT and IVGTT were decreased in subjects with IGT to levels comparable with those in NIDDM patients, whereas the response to GST was preserved in the subjects with IGT compared with NIDDM patients. The insulinogenic index of OGTT correlated well (r = 0.78) with early-phase insulin response to IVGTT, suggesting that the insulinogenic index in OGTT is related to the early-phase insulin response to IVGTT in nonobese subjects. CONCLUSIONS Impaired early-phase insulin response to glucose was associated with mild glucose intolerance, suggesting the importance of impaired insulin secretion in the development of glucose intolerance in nonobese subjects.

Suppression of synthesis and release of glucagon by glucagon-like peptide-1 (7–36 amide) without affect on mRNA level in isolated rat islets

Glucagon-like peptide-1 (GLP-1) has been reported to inhibit glucagon release. To investigate the mechanism involved, we examined the effects of GLP-1 on the preproglucagon mRNA level and the content and release of glucagon in the isolated rat islets. Arginine significantly increased the content and release of glucagon after incubation for 1 h or 18 h. The preproglucagon mRNA level was not increased after incubation for 1 h but was increased after incubation for 18 h. GLP-1 (7-36 amide) significantly decreased the content and release of glucagon after incubation for 1 h or 18 h, but did not affect arginine-induced increase in the preproglucagon mRNA level after incubation for 18 h. These data suggest that GLP-1 suppresses post-transcriptional processes in the content and release of glucagon.

Analysis of the method for conversion between levels of HbA1c and glycated albumin by linear regression analysis using a measurement error model

AIM To establish a method for conversion between HbA(1c) and glycated albumin (GA) using a measurement error model (MEM). METHODS Type 2 diabetic patients, without complications that might affect either HbA(1c) or GA, were enrolled in the study (n=154, age 68.4+/-9.9). HbA(1c), GA and postprandial plasma glucose (PPG) levels were measured simultaneously on >or=3 occasions. RESULTS PPG showed a significant correlation with HbA(1c) and GA (p<0.001 for both). Correlation between HbA(1c) and GA was very high (r=0.747, p<0.001). When the independent variable was assumed to be GA, common regression analysis yielded a regression line HbA(1c)=2.59+0.204GA. When the independent variable was changed to HbA(1c), the regression line became GA=2.26+2.74HbA(1c). The y-intercept of the first line was significantly positive, whereas that of the second was not. The regression line using MEM was HbA(1c)=1.73+0.245GA. The y-intercept was 1.73+/-0.38 (p<0.001) and the slope was 0.245+/-0.018 (p<0.001), showing that 1% increase in HbA(1c) level corresponds to 4% increase in GA level. CONCLUSIONS The relationship between HbA(1c) and GA was examined by regression analysis using MEM. HbA(1c) levels in Japan appear to have a positive shift of approximately 1.7%. Incremental ratio 4 of GA vs. HbA(1c) showed good consistency with values derived from in vitro data.

Renal handling of glycated albumin in non-insulin-dependent diabetes mellitus with nephropathy

Renal handling of glycated albumin in diabetic nephropathy was examined by studies on renal selectivity for glycated albumin in 23 normal controls and 52 patients with non-insulin-dependent diabetes mellitus (NIDDM) with various degrees of nephropathy. The serum and urinary levels of glycated albumin were measured by enzyme-immunoassay with monoclonal antibody to glucitol-lysine residues in human glycated albumin. The diabetic patients were divided into 3 groups according to the albumin index (AI): patients with normoalbuminuria [AI less than or equal to 30 mg/g creatinine(Cr)], with microalbuminuria (30 less than AI less than or equal to 270 mg/g Cr), and with macroalbuminuria (AI greater than 270 mg/g Cr). The renal selectivity for glycated albumin was calculated from the ratio of the urinary to serum level of glycated albumin. In the controls, the renal selectivity was as high as 4.40 +/- 0.48, and significantly higher than those in patients with normo- (2.87 +/- 0.29), micro- (1.72 +/- 0.20) and macroalbuminuria (1.26 +/- 0.23). The renal selectivity was inversely correlated with the AI in diabetic patients (r = -0.58, P less than 0.01). These data indicate that glycated albumin was selectively excreted in the urine and that the renal selectivity in diabetic patients gradually decreased to a value of 1 with increase in albuminuria. When the patients with normoalbuminuria were divided into two subgroups with high and low albumin excretion, the renal selectivities for glycated albumin in both subgroups were still significantly lower than that in controls. These results suggested that early diabetic nephropathy which cannot be detected clinically by albuminuria can be diagnosed by measurement of renal selectivity for glycated albumin.

Effects of very-low-calorie diet weight reduction on glucose tolerance, insulin secretion, and insulin resistance in obese non-insulin-dependent diabetics

We put 12 obese subjects on a very-low-calorie diet (VLCD) and observed how their weight loss affected their glucose tolerance. Seven had non-insulin-dependent diabetes and five did not. They consumed 1000 kcal/day for at least 1 week, then 420 kcal/day for 4 weeks, and 1000 kcal/day thereafter. VLCD improved glucose tolerance and insulin response to a glucose load in the diabetics and did not affect these parameters in the non-diabetics. It did not change insulin responsiveness to intravenous glucagon in either group. Both groups showed improved insulin resistance, as measured by an insulin suppression test. Regression analysis showed that insulin resistance correlates well with obesity and glycemic control. Weight reduction did not change hepatic insulin extraction. Thus, the improvement in glucose tolerance by some of the diabetics seems to have arisen from improvements in their insulin resistance and insulin response to a glucose load. Insulin resistance improved because of weight reduction and subsequent improvements in glycemic control.

Salivary gland glucagon is a fictitious substance due to tracer-degrading activity resistant to protease inhibitors

A high level of glucagon immunoreactivity was apparently detected in acid-saline extract from rat submandibular glands, but tracer glucagon added to the assay mixture was mostly damaged in spite of the presence of protease inhibitors commonly used in radioimmunoassay. Gel-filtration of the extract on a Bio-Gel P-10 column revealed strong tracer-degrading activity at the void fraction where the apparent immunoreactivity was eluted. Serial changes in apparent immunoreactivity of the extract fit well on the theoretical curve of an exponential tracer degradation. These findings indicate that the salivary gland glucagon is a fictitious substance due to tracer degradation during radioimmunoassay. Further study revealed that the glucagon molecule was hydrolyzed at the arginyl bonds and split into two fragments during incubation with the acid-saline extract from rat submandibular glands.

Aspartic acid at position 57 of the HLA-DQβ chain is not protective against insulin-dependent diabetes mellitus in Japanese people

Insulin-dependent diabetes mellitus (IDDM) in Caucasians is closely associated with the HLA-DQ gene, especially the residue 57 of the DQ beta chain. Aspartic acid at this position provides protection against IDDM, and substitution of this residue by alanine, valine or serine increases susceptibility to IDDM. To determine whether this is a common feature of IDDM in different ethnic groups, we studied DQB1 DNA in Japanese patients with IDDM by polymerase chain reaction and non-radioactive restriction site analysis. In contrast to Caucasian patients with IDDM, most Japanese patients with IDDM possessed at least one aspartic acid at position 57 of DQ beta. This finding strongly suggests that aspartic acid at position 57 of DQ beta does not protect the Japanese from IDDM.