Clinical and Investigative Medicine

 

Variable glycation of serum proteins in patients with diabetes mellitus

Marielle Couturier, MSc
Hélène Amman, PhD
Christine Des Rosiers, PhD
Ronald Comtois, MD

Clin Invest Med 1997;20(2):103-109.

[résumé]

Ms. Couturier and Dr. Des Rosiers are with the Department of Nutrition, University of Montreal; Dr. Amman was with the Department of Biochemistry, Notre-Dame Hospital; Dr. Comtois is with the Division of Endocrinology and the Department of Medicine, Notre-Dame Hospital, and the Faculty of Medicine, University of Montreal, Montreal, Que.

(Original manuscript submitted Jan. 5, 1995; received in revised form Dec. 14, 1996; accepted Jan. 11, 1997)

Reprint requests to: Dr. Ronald Comtois, Notre-Dame Hospital, 1560 Sherbrooke St. E, Montreal QC H2L 4M1; fax 514 876-7282

Contents
Abstract

Objective: To determine whether there were variations in vivo and in vitro in the glycation process among patients with diabetes mellitus and to assess the characteristics of patients with high and low glycation, if this was observed.

Patients: Patients (n = 185) attending a Diabetes Day Care Centre or Notre-Dame Hospital in Montreal participated in the in vivo study. Patients found to have high and low glycation were asked to allow the use of their serum for the in vitro part of the study.

Intervention: Capillary blood glucose levels were determined by nursing staff 4 times a day over 7.3 (standard deviation [SD] 5.3) consecutive days with commercially available glucose oxydase reagent strips and meters. The ratio of the fructosamine concentration to the protein concentration (the F/P ratio) and the glycated hemoglobin were also determined at the same time as the capillary blood glucose level. Glycation was defined as the mean capillary blood glucose/F/P ratio. Patients with high and low glycation (higher or lower than 1 SD of the mean) were compared. For the in vitro study, incorporation of carbon-14 glucose in serum proteins incubated with a 30-mmol/L glucose concentration was studied in some of the patients with low and high glycation.

Results: The mean capillary blood glucose/F/P ratio was a mean of 2.30 (SD 0.29) g/mL. Of the 185 subjects, 31 had high glycation (1.46 [SD 0.19] g/mL) and 27 had low glycation (2.97 [SD 0.035] g/mL, p < 0.001). There was no significant difference in age, sex, diabetic treatment and glycated hemoglobin levels between the 2 groups. However, patients with low glycation had a greater body mass index (29.4 [SD 5.7] kg/m2 v. 26.4 [SD 4.3] kg/m2, p < 0.05). In vitro, incorporation of 14C glucose in serum proteins incubated with a 30-mmol/L glucose concentration was higher in the serum of the 9 patients with high glycation than in that of the 7 with low glycation (0.031% [SD 0.03%] per gram of proteins v. 0.028% [SD 0.03%] per gram of proteins, p < 0.02).

Conclusions: Glycation may vary among patients with diabetes mellitus who have similar capillary blood glucose concentrations. Glycation appears to be lower in patients with a greater body mass index. Furthermore, alteration in the glycation process itself may explain, in addition to the mean blood glucose level, the difference in fructosamine levels.

Résumé

Objectif : Déterminer si le processus de glycation chez les patients atteints de diabète sucré varie in vivo et in vitro et évaluer les caractéristiques des patients à glycation élevée et à glycation faible dans les cas où l'on a observé le phénomène.

Patients : Des patients (n = 185) fréquentant un centre de jour pour diabétiques ou l'hôpital Notre-Dame à Montréal ont participé à l'étude in vivo. On a demandé aux patients chez lesquels on a constaté une glycation forte et une glycation faible l'autorisation d'utiliser leur sérum pour le volet in vitro de l'étude.

Intervention : Le personnel infirmier a déterminé la glycémie capillaire 4 fois par jour pendant une période de 7.3 (écart type [ET] 5.3) jours consécutives en utilisant des bandelettes de glucose-oxydase et des appareils de mesure disponibles sur le marché. On a aussi établi, en même temps que la glycémie capillaire, le ratio entre la concentration de fructosamine sur celle de protéine (le ratio F/P) et l'hémoglobine glycosylée. On a défini la glycation comme étant la glycémie capillaire moyenne/F/P. On a comparé les patients à glycation élevée et à glycation faible (supérieure ou inférieure d'un ET par rapport à la moyenne). Dans le cas du volet in vitro, on a étudié chez certains des patients à glycation faible et à glycation élevée l'incorporation de glucose au carbone-14 dans des protéines sériques incubées dans du glucose à 30 mmol/L.

Résultats : La glycémie capillaire moyenne/F/P s'est établi en moyenne à 2,30 (ET 0,29) g/mL. Sur les 185 sujets, 31 avaient une glycation élevée (1,46 [ET 0,19] g/mL) et 27, une glycation faible (2,97 [ET 0,035] g/mL, p < 0,001). Il n'y avait pas de différence significative quant à l'âge, au sexe, au traitement du diabète et aux taux d'hémoglobine glycosylée entre les 2 groupes. Cependant, les patients à glycation faible avaient un indice de masse corporelle plus élevé (29,4 [ET 5,7] kg/m2 c. 26,4 [ET 4,3] kg/m2, p < 0,05). Au cours do volet in vitro, l'incorporation de glucose au carbone 14 dans les protéines sériques incubées dans du glucose à 30 mmol/L a été plus forte dans le sérum des 9 patients à glycation élevée que dans celui des 7 patients à glycation faible (0,031 % [ET 0,03 %] par gramme de protéines c. 0,028 % [ET 0,03 %/g] par gramme de protéines, p < 0,02).

Conclusion : La glycation peut varier entre les patients atteints de diabète sucré qui ont des glycémies capillaires semblables. La glycation semble moins élevée chez les patients qui ont un indice de masse corporelle plus élevé. De plus, l'altération du processus de glycation peut expliquer en soi, outre la glycémie moyenne, des taux de fructosamine différents.

[ Top of document ]

Introduction

Assessment of glycemic control has always posed a problem in the management and care of patients with diabetes mellitus. Measurement of the capillary blood glucose level is reliable only if performed frequently. Furthermore, both the patient's performance and reporting of these tests may not be truly objective.

Fructosamine is a ketoamine formed by spontaneous nonenzymatic condensation of glucose and proteins. Its measurement provides a valuable retrospective index of the mean prevailing glucose level during a time interval ranging from 7 to 21 days, depending on the life span of the serum proteins.[1] The clinical usefulness of fructosamine assay in evaluating glycemic control is still being investigated. Alterations in serum fructosamine levels have been described in some pathologic states such as hypothyroidism, chronic renal disease and cirrhosis of the liver.[2­4]

The aims of this study were to determine whether there were variations in vivo and in vitro in the glycation process among patients with diabetes mellitus and to assess the characteristics of patients with high and low glycation, if these conditions were observed.

[ Top of document ]

Patients and methods

We studied 185 patients with diabetes mellitus attending a Diabetes Day Care Centre (n = 152) or Notre-Dame Hospital (n = 33) in Montreal. A history and an evaluation were conducted for each patient. Nutritional intake was evaluated by a registered dietician with the use of the frequency-recall method. The capillary blood glucose level was determined by the nursing staff 4 times a day (fasting, before lunch, before dinner and at bedtime) during 7.3 (SD 5.3) consecutive days with commercially available glucose oxydase reagent strips and meters (One Touch II, Lifescan Canada Ltd., Vancouver). Quality control of this meter is performed each month. The meter's interassay coefficient of variation (CV) is always less than 3.5% and its CV between nurses is less than 8.5%.

The ratio of the fructosamine concentration to the protein concentration (F/P ratio) and the glycated hemoglobin (GHb) were determined during the corresponding period.

Of the 185 patients included in the in vivo study, 58 were considered eligible to participate in the in vitro study. We wished to compare the 31 patients with high glycation (defined as a mean capillary blood glucose/F/P ratio lower than 1 standard derivation [SD] of the mean) and 27 with low glycation (defined as a mean capillary blood glucose/F/P ratio higher than 1 SD of the mean). We were unable to contact 23 of the eligible patients, and 19 refused to participate. Consequently, we performed the in vitro studies with the use of serum from 9 patients with high glycation and 7 with low glycation.

A test kit (Fructosamine Test Plus, Roche, Switzerland) was used to determine fasting serum fructosamine concentrations obtained on the last day of the capillary glucose determination. The fructosamine interassay CV was 1.9%.[5] Total protein concentrations were determined by a Buiset reaction rate method, and the F/P ratio was calculated. The normal range for the F/P ratio is 2.9 to 4.0 µmol/g. GHb was determined by affinity chromatography (Glyc-Affin system, Isolab Inc., Ohio) with the use of disposable mini-columns filled with phenylboronic acid. The interassay CV for GHb determinations was 5.1%, which is within the normal range of 4.9% to 6.6%.

Glycation of serum proteins was investigated in vitro with the use of a protocol by Broussolle and associates.[6] Subsequently, 2 mL of serum were incubated with 1 µCi of carbon 14 glucose (Amersham 230 Ci/mol) for 24 hours at 37°C in a humidity-controlled atmosphere containing 5% carbon dioxide to maintain a pH of 7.3 to 7.4. Cold glucose was added to achieve a final glucose concentration of 30 mmol/L. A 24-hour incubation time was chosen, but the incorporation of 14C glucose into acid-precipitable fraction of human serum has been shown to increase linearly with time between 1 and 9 days.[6] Before and after the 24 hours' incubation, 0.1 mL of the incubated proteins was removed. The proteins were precipitated with 1.0 mL of cold 10% (w/v) trichloroacetic acid. The precipitate was dissolved in 0.2 mL of sodium hydroxide, and the procedure was repeated. The final precipitate was dissolved in 10 mL of liquid scintillation counter. The incorporation of 14C glucose was measured by subtracting the level of radioactivity in the sample before incubation from that afterward.

[ Top of document ]

Data analysis

The statistical analyses were performed with the Mann­Whitney U test. Linear regression analyses were conducted and correlation coefficients determined.

[ Top of document ]

Results

We observed a good correlation (r = 0.59) between the mean capillary blood glucose level and the F/P ratio at corresponding times among the 185 patients with diabetes mellitus (Fig. 1). The regression model accounted for 34.8% of the variation observed. However, there were occasionally wide variations in F/P ratios for similar mean capillary blood glucose concentrations. The correlation between the patient's body mass index (BMI) and the F/P ratio was -0.15 (p < 0.05), and the BMI accounted for 2.3% of the variation. There was no significant difference in F/P ratio according to age, sex or type of diabetes mellitus. Patients with insulin-dependent diabetes mellitus (IDDM) had a lower BMI than those with non-insulin-dependent diabetes mellitus (NIDDM) (23.6, SD 2.4 kg/m2 v. 29.3, SD 5.1 kg/m2, p < 0.001). The BMI was similar among male and female patients.

In the in vivo study, we compared the 31 patients with high glycation and the 27 with low glycation. There was no significant difference between the 2 groups of patients with regard to age, sex and type of diabetes mellitus (Table 1). In addition, there was no significant difference in regard to the incidence of microvascular and macrovascular complications or polyneuropathy. However, patients with low glycation had a higher BMI (p < 0.05). After excluding patients with IDDM in both groups, the patients with low glycation who remained still had a signficantly higher BMI (p < 0.03). The correlation between BMI and the mean blood glucose/F/P ratio (glycation) was 0.25 (p = 0.06). However, the F/P ratio was similar in obese (BMI >= 30 kg/m2) and leaner patients (BMI < 30 kg/m2) (4.80 v. 5.02 µmol/g, p = 0.22). The higher mean glucose levels in obese patients could explain the lack of difference (12.8, SD 2.2 mmol/L v. 8.3, SD 2.1 mmol/L). Comparison of the biochemical characteristics of the 2 groups (Table 2) revealed that GHb was similar in patients with high and low glycation. However, the mean capillary blood glucose level was lower and the F/P ratio higher in the group of patients with high glycation. Consequently, the mean capillary blood glucose/F/P ratio was significantly lower in this group.

The clinical and biochemical characteristics of the 9 patients with high glycation and the 7 with low glycation in whom we performed the in vitro study were similar to those of other patients with high and low glycation, respectively (Tables 1 and 2). In the in vitro study, the glycation rate was assessed by measuring the incorporated radioactivity of 14C glucose in the serum proteins after 24 hours of incubation with glucose concentrations of 30 mmol/L. The resulting rate was higher in the serum of patients with high glycation than in the serum of those with low glycation (0.32%, SD 0.03% per gram of proteins v. 0.028%, SD 0.003% per gram of proteins; p < 0.02) (Fig. 2).

[ Top of document ]

Discussion

Serum fructosamine levels are related to nonenzymatically glycated serum protein levels and are considered to reflect mean blood glucose changes during the past 2 to 3 weeks.[1] Several factors are known to influence the glycation of proteins in vivo and in vitro. In vivo, the concentration of fructosamine is related to the mean blood glucose level. High and significant correlations (r = 0.53 to 0.71) between mean capillary blood glucose concentrations and serum fructosamine levels in patients with diabetes have been reported.[7,8] Our data support these observations, showing a correlation coefficient of 0.59.

Various observations have been made regarding the influence of the total protein level on the fructosamine level. Some authors have found a high correlation between the 2 variables, whereas others have stated that the fructosamine concentration is independent of the total protein concentration.[9­12] Desjarlais and associates[5] observed that there was a better correlation between the F/P ratio and GHb than between the fructosamine level and GHb. Therefore, we have suggested that the total protein concentration and the F/P ratio be reported with the fructosamine result.[5]

Alterations of serum fructosamine have been observed in clinical situations when the turnover of proteins is affected. Thus, fructosamine concentrations have been found to be increased in hypothyroidism and decreased in hyperthyroidism.[2,3] In addition, Jensen and Haymond[13] have observed an increased protein turnover in obese women. Indeed, some studies have also reported low fructosamine levels in obese patients.[6,14­16] These authors observed that the rate of formation of fructosamine and the incorporation of 14C glucose into serum proteins were decreased in obese subjects.[6,16] The radioactivity incorporation into proteins increased linearly with time between 1 and 9 days for both obese and control subjects, but a significantly lower incorporation was observed for obese subjects at all times, despite variations between incubations in the final glucose concentration and protein content. In this study, the final glucose concentration was standardized to 30 mmol/L.

In the presence of abnormal proteins, such as high IgA levels, serum concentration of fructosamine has been found to be increased.[17] Indeed, serum immunoglobulins have a higher tendency for nonenzymatic glycation and, thus, any variation in these proteins may influence serum fructosamine levels. In the presence of multiple myeloma IgG, a decrease of the fructosamine concentration was observed.[18] However, after correcting the value for albumin, the difference disappeared.

Fructosamine is formed by spontaneous, nonenzymatic condensation of glucose and the amino groups of proteins. The initial product is an unstable aldimine, the Schiff base, which, by the amadori rearrangement, converts to a stable ketoamine.[1] In vitro, there are many factors that might influence the extent of nonenzymatic glycation: glucose concentration, incubation time, temperature, pH, protein concentration and the microenvironment of the proteins. The glucose concentration, the incubation time, the pH and the temperature were, in vitro, identical in the 2 groups in our study. In addition, we expressed the results as the F/P ratio to correct for the influence of the protein concentration on glycation.

It has been suggested that some factors in the microenvironment of the proteins might affect the kinetics and specificity of protein glycation. For instance, Watkins and associates[19] observed that the glycation of hemoglobin was increased in the presence of phosphate. Hence, differences in the proteins' local environment might explain, at least in part, the different rates of glycation observed in our 2 groups. Obese patients may have a different microenvironment for protein. Nonenzymatic glycation is inhibited by dibasic amino acids arginine and lysine in rat-tail tendons, according to Menzel and Reihsner.[20] One may speculate that levels of such amino acids are higher in obese than nonobese patients with diabetes mellitus, which may explain the lower glycation rate in obese subjects. Other factors, such as free fatty-acid concentration, which is increased in obese patients, may be involved.

There is an increasing body of evidence indicating that glycation products are major factors in the development of micro- and macrovascular complications of diabetes mellitus. Nonenzymatic glycation results from a direct chemical reaction between glucose and amino groups on proteins. The first stable products of the reactions are termed ketoamines. They are known to undergo further dehydration, rearrangement and cleavage reactions. The final products are fluorescent and highly cross-linked; these are called advanced glycation end products (AGEs). Makita and associates[21] observed that AGEs accumulate at an accelerated rate in arterial vessel walls and in the circulation of patients with diabetes mellitus, and that the increase in circulating AGE peptides parallels the severity of renal function impairment in diabetic nephropathy.

[ Top of document ]

Conclusion

Glycation may vary from one patient with diabetes mellitus to another, even for patients with the same mean capillary blood glucose level. It appears to be lower in patients with a greater BMI. In vitro, incorporation of 14C glucose in serum proteins was higher in the serum of patients with clinically higher glycation. Hence, in addition to the mean blood glucose level, the difference in fructosamine levels might be explained by alteration in the glycation process itself.

[ Top of document ]

Acknowledgements

We are grateful to Joanne Auclair for providing excellent secretarial support.

[ Top of document ]

References

  1. Johnson RN, Metcalf PA, Baker JR. Fructosamine: a new approach to the estimation of serum glycosylprotein. An index of diabetic control. Clin Chim Acta 1982;127:87-95.
  2. Cirillo R, Balgano S, Cossu E, et al. The effect of altered thyroid function on serum fructosamine concentrations. Clin Biochem 1988;21:179-81.
  3. Ford HC, Chew-Lim W, Crooke MJ. Hemoglobin A1C and serum fructosamine levels in hyperthyroidism. Clin Chim Acta 1987;166:317-21.
  4. Constanti C, Simo JM, Joven J, Camps J. Serum fructosamine concentration in patients with nephrotic syndrome and with cirrhosis of the liver: the influence of hypoalbuminaemia and hypergammaglobulinaemia. Ann Clin Biochem 1992;29:437-42.
  5. Desjarlais F, Comtois R, Beauregard H, et al. Technical and clinical evaluation of fructosamine determination in serum. Clin Biochem 1989;22:329-35.
  6. Broussolle C, Tricot F, Garcia I, et al. Evaluation of the fructosamine test in obesity: consequences for the assessment of past glycemic control in diabetes. Clin Biochem 1991;24:203-9.
  7. Cefalu W, Parker T, Johnson C. Validity of serum fructosamine as index of short-term glycemic control in diabetic outpatients. Diabetes Care 1988;11:662-4.
  8. Comtois R, Desjarlais F, Nguyen M, Beauregard H. Clinical usefulness of estimation of serum fructosamine concentration as screening test for gestational diabetes. Am J Obstet Gynecol 1989;160:651-4.
  9. Baker JR, O'Connor JP, Metcalf PH, et al. Clinical usefulness of estimation of serum fructosamine as a screening test for diabetes mellitus. BMJ 1983;287:863-7.
  10. Fluckiger R, Woodli T, Berger W. Evaluation of the fructosamine test for the measurement of plasma protein glycation. Diabetologia 1987;30:648-52.
  11. Mosca A, Carenini A, Zoppi F, et al. Plasma protein glycation as measured by fructosamine assay. Clin Chem 1987;33:1141-6.
  12. Van Dieijen-Visser MP, Seynaeve C, Brombacher PJ. Influence of variations in albumin or total-protein concentration on serum fructosamine concentration. Clin Chem 1986;32:1610.
  13. Jensen MD, Haymond MW. Protein metabolism in obesity: effects of body fat distribution and hyperinsulinemia on leucine turnover. Am J Clin Nutr 1991;53:172-6.
  14. Skrha J, Svacina S. Serum fructosamine and obesity [letter]. Clin Chem 1991;37:2020-1.
  15. Woo J, Cockram C, Lau E, et al. Influence of obesity on plasma fructosamine concentration. Clin Chem 1992;38:2190-2.
  16. Ardawri MSM, Nasrat HAN, Bahnassy AA. Fructosamine in obese normal subjects and type 2 diabetes. Diabet Med 1994;11:50-6.
  17. Montagna M, Laghi F, Cremona G, et al. Influence of serum proteins on fructosamine concentration in multiple myeloma. Clin Chim Acta 1991;204:123-30.
  18. Comtois R, Couturier M, Amman H. Influence of protein and albumin levels on serum fructosamine concentration in a diabetic patient with multiple myeloma. Clin Biochem 1994;27:421-3.
  19. Watkins N, Neglia-Fisher C, Dyer D, et al. Effect of phosphate on the kinetics and specificity of glycation of protein. J Biol Chem 1987;262:l7207-12.
  20. Menzel EJ, Reihsner R. Alterations of biochemical and biochemical properties of rat tail tendons caused by non-enzymatic glycation and their inhibition by dibasic amino acids arginine and lysine. Diabetologia 1991;34:12-6.
  21. Makita Z, Radoff S, Rayfield EJ, et al. Advanced glycation end products in patients with diabetic nephropathy. N Engl J Med 1991;325:836-42.

| CIM: April 1997 / MCE : avril 1997 |
CMA Webspinners / >