- Email: [email protected]
Increased accumulation of nonenzymatically glycated fibrinogen in the renal cortex in rats
THROMBOSIS RESEARCH 58; 23-33,199O 0049-3848/90 $3.00 + .OO Printed in the USA. Copyright (c) 1990 Pergamon Press plc. All rights reserved.
INCREASED ACCUMULATION OF NONENZYMATICALLY GLYCATED* FIBRINOGEN IN THE RENAL CORTEX IN RATS.
Takashi Murakami, Hiroshi Egawa, Yutaka komiyama, Midori Masuda, and Kenjiro Murata Department of Clinico-laboratory Medicine, Kansai Medical University, Moriguchi, Osaka 570, JAPAN (Received 19.10.1989; accepted in revised form 16.1 .1990 by Editor S. Okamoto)
ABSTRACT We have determined that the nonenzymatic glycation of fibrinogen altered its biological functions in vitro. Thrombin clottability of rat fibrinogen incubated with glucose decreased with increasing incubation time, but was not affected by the glucose concentration. Fibrin prepared from glycated fibrinogen showed a significant resistance in susceptibility to plasmin degradation. We also examined the in vivo distribution of glycated fibrinogen in renal cortex. Iodine labeled rat glycated or unglycated fibrinogen was injected into streptozocin-induced diabetic and control rats. NO appreciable difference in the plasma disappearance rate in control rats was observed (half-lives in hours for glycated, 25.6 f 0.37; unglycated, 26.1 k 0.74). The radioactivity of fibrinogen retained to the renal cortex was calculated 2,!+-hours after injection. In the retention rate of glycated both control and diabetic rats, fibrinogen in renal cortex was significantly higher than that of the unglycated. These results suggest that glycated fibrinogen may fibrin occur in a more resistant form to plasmin digestion with deposition as confirmed in in vitro studies. Therefore, we suggest that glycated fibrinogen may partly contribute to the development of diabetic microangiopathic lesions such as glomerulosclerosis.
Key words : Nonenzymatic glycation (Glycosylation), Rat fibrinogen, Diabetic microangiopathies,
Fibrin deposition.
*The Joint Commission on Biochemical Nomenclature of the International Union of Biochemistry (IUB) and the International Union of Pure and Applied Chemistry (IUPAC) suggests the term glycation rather than glycosylation (or glucosylation) for the nonenzymatic reaction that links a glucose or other sugars to a protein (6). 23
24
RAT GLYCATED FIBRINOGEN IN VIVO
Vol. 58, No. 1
INTRODUCTION Fibrin(ogen) is one of several proteins that have been immunohistochemically demonstrated in diabetic angiopathic lesions, especially in renal glomerular basement membranes (GBM) and mesangium (1). In the glomerus, the deposition of these proteins causes increasing permeability and expanding Finally mesangium, and leads to secondary accumulation of proteins. capillary occlusion and nodular droplets occur. On the other hand, in diabetic patients, various abnormalities in coagulation and fibrinolysis are found (2-5), i.e., higher fibrinogen concentrations, elongation of the thrombin clotting time, reduced fibrinogen survival, and elevated fibrinopeptide A and fibrin degradation products. It is not known whether these abnormalities are due to fibrinogen itself. Because some of these abnormalities are reversed by appropriate treatment for hyperglycemia (3,4), there is a possibility that fibrinogen alters its biological function during hyperglycemia. Recently it has been reported that nonenzymatic glycation (6) of fibrinogen increased in poorly controlled diabetic patients as well as other proteins (7-IO), and that glycated fibrinogen modified its native biological functions in vitro (11-14). Therefore, we hypothesized that fibrin deposition as seen in diabetic microangiopathic lesions is due to increased glycated fibrinogen which alters its native biological functions. In the present studies, we injected labeled glycated fibrinogen into streptozocin (STZ) induced diabetic and control rats. We compared plasma half-life and renal cortical accumulation of glycated fibrinogen with those of the unglycated.
MATERIALS AND METHODS Animals Male Sprague-Dawley rats, 9 weeks old and weighing 230-250 g, were purchased from Japan SLC. (Hamamatsu, Japan). Diabetes was induced by intravenous injection of 25 mg/kg STZ (lot 605527, Calbiochem, La Jolla, CA), which was freshly dissolved in 0.9% saline acidified with 10 mM citric acid pH 4.2. Diabetes was confirmed by the presence of hyperglycemia, which was checked once a week for 3 months after the administration of STZ. Animals with nonfasting plasma glucose levels of less than 360 mg/dl were rejected. Diabetic rats and age-matched control rats were allowed free access to food and water. In vivo experiments were performed after 3 months of maintenance. Glycation of fibrinogen Fibrinogen from fresh rat plasma was purified by cold ethanol precipitation (15), and contaminating plasminogen was removed by chromatography on lysine-Sepharose (Pharmacia, Uppsala, Sweden) by the method of Matsuda et a1.(16). This purified fibrinogen (92% clottable) was dissolved in 50 mM sodium phosphate buffer pH 7.4 with 0.15 M sodium chloride (PBS), and glucose added to this. The solutions were sterilized by passing through a 0.45 urnMillipore filter and then incubated for up to 96 hours at 37°C. Free glucose was removed by dialysis extensively at 4°C against PBS. Quantification of glycated fibrinogen was determined by affinity chromatography by means of a m-aminophenylboronic acid-bound column
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN VIVO
25
(Glyco-Gel B; Pierce, Rockford, IL) as previously described (IO, 17). Briefly, fibrinogen solution was dialyzed against 0.25 M sodium ammonium buffer pH 8.5, and then applied to a column which was pre-equilibrated with the same buffer. After collecting the unglycated fraction passed through, the bound glycated fibrinogen was eluted with 0.25 M sodium acetate buffer The percentage of glycation (%glycation) was measured by pH 5.5. calculating the total absorbance at 280 nm (Aneo) of each fibrinogen fraction. Biological function of the glycated fibrinogen in vitro Thrombin clottability of incubated fibrinogen was measured by the method of Laki (18). For estimating the degradation of fibrin formed from glycated fibrinogen (glycated fibrin) by plasmin digestion, a modified fibrin plate assay (19) was performed. Briefly, the glycated (incubated with 500 mM glucose for 72 hours) and unglycated (without glucose, 72 hours) fibrinogen was made by dissolving l+mg/ml in PBS. Agarose (Agarose L; low gel point type, Wako Pure Chemicals, Osaka, Japan) was dissolved in PBS containing 0.02% sodium azide. Three ml of fibrinogen solution and 7 ml of agarose solution were mixed in each plastic plate at /+2"C, and allowed to cool to room temperature. After setting to gel, the plates were clotted by adding 3 ml of 10 U/ml of thrombin and incubated at 37°C for 2 hours. Exactly 3 ~1 of different concentrations (1, 6 casein units (ClJ)/ml)of standard plasmin solution (KabiVitrum, Stockholm, Sweden) was placed in a The well which was holed in the surface of the fibrin plates. by plasmin was determined by susceptibility of fibrin to degradation measuring the mean of two perpendicular diameters of the digested fibrin area. Iodine labelling The glycated fibrinogen used in the following experiments was made by incubation with 500 mM glucose for 72 hours, and was separated from unglycated fractions by borate affinity chromatography. Glycated and unglycated fibrinogen were iodinated using Na1251 (Du Pont Company, Wilmington, DE) and 1,3,4,6-tetrachloro-3a,6a-diphenylglycouril(IODOGEN ; Pierce) by the method of Nieuwenhuizen (20). Assessment of clottable radioactivity in the labeled fibrinogen was performed according to Blomback et al. (21). The final specific radioactivities and clottable activities were 6.6 x 10' cpmlmg, 86.2% of glycated, and 6.7 x IO7 cpm/mg, 88.0% of unglycated While iodination was carried out after glycation, it was fibrinogen. possible that the conditions of iodination might modify the nature of the glucosyl adduct (22), so we checked whether the glycated fibrinogen still had affinity to the boronate-Sepharose column after iodination (data not shown). These were stored at -70°C until use. The circulating half-lives of labeled fibrinogens Control rats (10 animals, weighing 240 + 7.5 g) were injected via the jugular vein with either 0.06 mg of 1251-glycated or '251-unglycated fibrinogen under mild ether anesthesia. A series of blood samples (0.4 ml each) was drawn from the opposite jugular vein with a heparinized syringe at 6 minutes and over a 6-day period. Radioactivity was determined in 0.1 ml aliquots of plasma precipitated with an equal volume of 20% trichloroacetic acid (TCA-precipitated plasma).
26
RAT GLYCATED FIBRINOGEN IN VW0
Vol. 58, No. 1
Accumulation of glycated fibrinogen in renal cortex in vivo Labeled fibrinogens were diluted with saline and final concentrations were 0.2& mg/ml. Control and diabetic rats were divided randomly into two subgroups. These animals were injected with either glycated or unglycated fibrinogen (1 ml per 1 kg rat body weight), and blood samples were drawn at 6 minutes. At 2f,-hoursafter injection, to remove the unbound fibrinogen Rats were from renal cortex, rat kidneys were perfused as follows. anesthetized with ether and the abdominal aorta exposed through an abdominal incision. The abdominal aorta and inferior vena cava were ligated just above the right renal veins, and kidneys were then perfused via the abdominal aorta and the renal arteries with 100 ml of saline containing 5 U/ml heparin sulfate. Kidneys were extirpated and weighed, then the medullae were excised. The radioactivity of these cortices was measured directly and expressed as counts per minute (cpm) per wet weight of renal cortices (g). The retention rate of labeled fibrinogen in renal cortex was given as percent of radioactivity in TCA-precipitated plasma at 6 minutes. Analytical procedures Blood glucose levels, blood urea nitrogen and serum creatinine concentrations were measured by an autoanalyzer (Beckman, Brea, CA). Plasma fibrinogen concentrations were measured by the thrombin time method. Protein concentration was determined by measuring Aneo (Al%/lcm 280nm = 15.4 (23) for rat fibrinogen). The radioactivity of samples was measured in a gamma spectrometer (Packard, Downers Grove, IL). Statistics Results are expressed as means + SE. Statistical methods used were Student's t-test for paired observations and linear-regression analysis with t-test of regression coefficient. P values reported were for one-tailed tests and values < 0.05 were considered significant.
RESULTS In
vitro
experiments
As shown in Fig. 1, the %glycation of fibrinogen was dependent on the length of incubation time and glucose concentration, but in 500 mM glucose it reached a plateau at 72 hours. The thrombin clottability of these fibrinogen decreased with time, but was not affected by the glucose concentration (Fig. 2). We also examined sodium dodecylsulfate polyacrylamide gel electrophoresis to determine the molecular degradation of each incubated fibrinogen, it showed identical patterns (data not shown). These results indicate that glucose, which covalently binds to the fibrinogen molecule, did not modify the thrombin clotting mechanism. Fibrin plate assay indicated the time course of plasmin digestion for glycated fibrin compared with that from the unglycated (Fig. 3), and it revealed that glycated fibrin was resistant to degradation by plasmin.
RAT GLYCATED FIBRINOGEN IN WV0
Vol. 58, No. 1
INCUBATION
TIME
27
( H )
FIG. 1. The %glycation of fibrinogen determined by using boronate affinity chromatography. Rat fibrinogen, 2.5 mg/ml of PBS, was incubated with each 20 mM (A), 50 mM (a), 100 mM (A,), 500 mM (0) glucose at 37°C. At indicated times the incubation mixture was dialyzed against 0.25 M ammonium acetate pH 8.5 at 4°C overnight, and 2 ml of this solution was applied to the column (bed volume: 4 ml). Each point represents the mean f SE of 3 separate experiments performed in duplicate.
100
X -
1
1
’
0
I
20 GLUCOSE
50 CONCENTRATION
100 (
500
mM 1
FIG. 2. Time course of thrombin clottability of rat fibrinogen incubated Fibrinogen, 2.5 mg/ml of PBS, was incubated with each with glucose. indicated concentration of glucose at 37°C for 24 (01, 48 (A,, 72 (A), 96 (0) hours. Each point represents the mean f SE.
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN WV0
12.5 -
. 6
12 INCUBATION
24 TIME
( H 1
FIG. 3. Effects of plasmin digestion with fibrin plates made from glycated and unglycated fibrinogen. Fibrinogen, 10 mg/ml of PBS, was incubated with or without 500 mM glucose at 37°C for 72 hours. Fibrin plates were prepared from glycated (0) and unglycated (0) fibrinogen, and were digested with 6 CU/ml (closed lines) and 1 CU/ml (broken lines) plasmin. Rate of digestion by plasmin was determined by measuring the mean of two perpendicular diameters of the digested fibrin area. Each point represents the mean + SE of 25 determinations. *P < 0.01 vs. unglycated.
In vivo
experiments
Laboratory and physiological data are shown in Table 1. Despite the loss of body weight in diabetic rats, the kidney weights were significantly greater than those of control rats. The plasma fibrinogen concentrations were significantly reduced in diabetic rats compared with control rats. The creatinine values were significantly increased in diabetic rats.
TABLE
1.
Characteristics of Experimental Animals at Time of Sacrifice Control (n = 20) Body weight (g) Kidney weight (g) Blood glucose (mg/dl) Fibrinogen (mg/dl) Blood urea nitrogen (mg/dl) Serum creatinine (mg/dl) Values are means f SE.
449.6 2.83 110.3 260.2 20.9 0.51
* + f * * *
3.0 0.28 4.5 6.2 0.46 0.018
N.S. = not significant.
Diabetic (n = 18) 307.5 3.08 518.1 185.8 22.2 0.68
f + * + f f
8.5 0.09 13.7 9.8 0.92 0.092
*P < < < <
0.01 0.05 0.01 0.01 N.S. < 0.05
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN WV0
29
In vivo experiments showing the kinetics of clearance of labeled fibrinogen were performed (Fig. 4). These radioactivity data between 24 and 124 hours possessed the linearity, and no significant difference in plasma disappearance rate for fibrinogen was found in the control rat circulation. The plasma radioactivity half-lives of glycated and unglycated fibrinogen were 25.6 * 0.4 hours and 26.1 + 0.7 hours, respectively. This result documents that glycation process does not affect the biological half-life of fibrinogen.
Fig. 5 shows the accumulation of labeled glycated and unglycated fibrinogen in renal cortex 24-hours after injection. In both control and diabetic rats, the retention rate of glycated fibrinogen in renal cortex was significantly higher than that of the unglycated. But each glycated and unglycated fibrinogen had lower retention rate in diabetic rats compared with that in control rats.
100 50
x -
10
c ; : s d g
1
f x
0.1
I
I
I
I
I
24
40
72
96
120
TIME
( H 1
Time course of disappearance of '251-glycated and unglycated FIG. 4. fibrinogen from the rat circulation. Control animals (five per experiment) were intravenously injected with either 0.06 mg of '251-glycated (0) or The disappearance of these labeled '*'I-unglycated (0) fibrinogen. fibrinogen from the circulation monitored in blood samples taken at Plasma fibrinogen radioactivities were expressed as indicated times. percentage of initial value, which was based on radioactivity at 6 minutes. Each point represents the mean f SE.
RAT GLYCATED FIBRINOGEN IN WV0
t
i-7
_I m
Vol. 58, No. 1
Glycated Unglycated
0
CONTROL
DIABETIC
Accumulation of '*'I-glycated and unglycated fibrinogen in renal FIG. 5. cortex. The retention rate was determined as described under MATERIALS AND METHODS. Each value represents the mean + SE. *P < 0.01 vs. unglycated.
DISCUSSION To prove the hypothesis that the glycated fibrinogen play an important role in the development of diabetic angiopathy, we have attempted to experimental demonstrate glycated fibrinogen in microvessels using caused animals. Repeated injections of glycated plasma proteins in mice GBM thickening and mesangial changes similar to those observed by electron microscopy in humans and experimentally induced diabetes (24). On the other hand, exogenous glycated albumin does not bind to GBM in diabetic or control However, these rats by means of immunofluorescence techniques (25). analytical methods make no distinction between injected proteins and endogenous protein deposition. If we can directly demonstrate in diabetic kidneys that the proportion of glycated fibrin(ogen) which accumulates in CBM and mesangial areas is greater than that which is present in plasma, it would enable us to prove that glycated fibrinogen contributes to the development of microangiopathic complications. But nonenzymatic glycated fibrin(ogen) is identified with unglycated fibrinogen by conventional immunohistochemical methods. So in the present studies, we used radiolabeled materials to reveal that glycated fibrinogen attached to renal cortex at higher concentrations than the unglycated. We demonstrated that the half-lives of radiolabeled glycated and unglycated fibrinogen in vivo were about l-day, which was compatible with results reported by Nieuwenhuizen et a1.(20). Glycation did not affect the biological half-life of fibrinogen as it did that of albumin (23). Since the clottabilities of these fibrinogens were equal, it could be speculated that the glycation process did not denature the fibrinogen molecule, and
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN WV0
31
also did not increase fibrinogen consumption. It further confirms the results of Jones et a.Z.(3)who reported that reduced fibrinogen survival in diabetes patients was not due to modified fibrinogen molecules. As the fibrin plate assay revealed that glycated fibrin was resistant to degradation by plasmin, glycated fibrinogen may be in a more insoluble form. This result confirmed a previous report of Brownlee et a1.(12). On the contrary, Ney et a1.(13) reported that the degradation of fibrin clots formed from glycated and unglycated fibrinogen were identical, but they used material which was incubated with a 20 mM glucose concentration. Our data using affinity chromatography indicated that the Xglycation under such conditions was maintained at a low level. It seems likely that differences in degradation rates were not observed because of little amounts of glycated fibrinogen in their materials. In both diabetic and control rats, we showed that the retention rate of glycated fibrinogen to renal cortex was still higher than that of the unglycated. But each glycated and unglycated fibrinogen had lower retention Although this rate in diabetic rats compared with that in control rats. reason is not clear, we suggest that some coagulation and fibrinolytic disorders which existed in diabetic states might interfere the fibrin deposition (26-28) or that the proliferation of endogenous fibrinogen binding to renal GBM during 3-months of STZ induced diabetes prevented the attachment of administered fibrinogen. In any case, we concluded that enhanced accumulation of glycated fibrin(ogen) in renal cortex was caused by the its resistance to degradation, and thus it seems to be evidence of initial phase of diabetic microangiopathic changes. Some investigators have assumed that glycated fibrinogen is physiologically not important because of the short half life of fibrinogen (humans, 3 - 4 days; rats, 1 day ) and the plasma glucose levels as low as 50 mM even in severe diabetic patients resulting in the glycated fibrinogen to appear in a small proportion in the circulation (13, 14). However, from our results, it seems that in chronic hyperglycemic states, a continuous exposure to glycated fibrinogen in the microcirculation might not be underestimated. Therefore, we suggest that glycated fibrinogen may partly contribute to the development of diabetic microangiopathic lesions such as glomerulosclerosis.
REFERENCES Diabetes mellitus and gout. In: Pathology of the 1. HEPTINSTALL, R.H. Kidney, 3rd edition. Little, Brown and Co., Boston, pp.139'7-1453,1983. Fibrinocoagulopathy in matu2. BANERJEE, R.N., SAHNI, A.L. and KHMAR, V. Thromb. Diath. and atherosclerosis. rity onset diabetes mellitus Haemorrh. 30, 123-132, 1973. Reduced fibrinogen survival in diabetes 3. JONES, R.L. and PETERSON, C.M. mellitus. A reversible phenomenon. J. Clin. Invest. 63, 485-493, 1979. Fibrinopeptide-A in diabetes mellitus. Relation to levels 4. JONES, R.L. of blood glucose, fibrinogen disappearance, and hemodynamic changes. Diabetes 34, 836-843, 1985.
32
RAT GLYCATED FIBRINOGEN IN WV0
Vol. 58, No. 1
Lower molecular weight fibrinogen and 5. GUREWICH, V. and LIPINSKA, I. non-clottable fibrinogen derivatives in diabetic mellitus. Thromb. Res. 22, 535-541, 1981. 6. NOMENCLATURE COMMITTEE OF IUB (NC-IuB) : IUB-IUPAC JOINT COMMISSION ON BIOCHEMICAL NOMENCLATURE (JCBN). Hoppe-Seyler's Z. Physiol. Chem. 365, I-V, 1984. Glycosyla7. LCTJENS, A., TE VELDE, A.A., VD VEEN, E.A. and VD MEER, J. Diabetologia 28, 87-89, 1985. tion of human fibrinogen in vivo. 8. KOENIG, R.J., PETERSON, C.M., JONES, R.L., SAUDEK, C., LEHMAN, M, and CERAMI, A. Correlation of glucose regulation and hemoglobin Ale in diabetes mellitus. N. Engl. J. Med. 295, 417-420, 1976. Nonenzymatically glucosylated 9. DAY, J.F., THORPE, S.R. and BAYNES, J.W. albumin. ; In vitro preparation and isolation from normal human serum. J. Biol. Chem. 254, 595-597, 1979. 10. KANESHIGE, H. Nonenzymatic glycosylation of serum IgG and its effect on antibody activity in patients with diabetes mellitus. Diabetes 36, 822828, 1987. 11. MCVERRY, B.A., THORPE, S., JOE, F., GAFFNEY, P. and HUEHNS, E.R. Nonenzymatic glucosylation of fibrinogen. Haemostasis 10, 261-270, 1981. 12. BROWNLEE, M., VLASSARA, H. and CERAMI, A. Nonenzymatic glycosylation reduces the susceptibility of fibrin to degradation by plasmin. Diabetes 32, 680-684, 1983. 13. NEY, K.A., PASQUA, J.J., COLLEY, K.J., GUTHROW, C.E. and PIZZO, S.V. In vitro preparation of nonenzymatically glucosylated human transferrin, a2-macroglobulin, and fibrinogen with preservation of function. Diabetes 34, 462-470, 1985. 14. MIRSHAHI, M., SORIA, J., SORIA, C., BERTRAND, O., MIRSHAHI, M. and BASDEVANT, A. Glycosylation of human fibrinogen and fibrin in vitro its consequences on the properties of fibrin(ogen). Thromb. Res. 48, 279-289, 1987. 15. DOOLITTLE, R.F., SCHUBERT, D. and SCHWARTZ, S.A. Amino acid sequence studies on artinodactyl fibrino-peptides. Biochem. Biophys. Acta. 118, 456-467, 1967. 16. MATSUDA, M., IWANAGA, S. and NAKAMURA, S. A simple, large scale method for preparation of plasminogen-free fibrinogen. Thromb. Res. 1, 619630, 1972. 17. MALLIA, A.K., HERMANSON, G.T., KROHN, R.I., FUJIMOTO, E.K. and SMITH, P.K. Preparation and use of a boronic acid affinity support for separation and quantitation of glycosylated hemoglobins. Anal. Lett. 14, 649-661, 1981. 18. LAKI, K. The polymerization of proteins. ; The action of thrombin on fibrinogen. Arch. Biochem. Biophys. 32, 317-324, 1951.
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN VIVO
33
The fibrin plate method for estimating 19. ASTRUP, T. and MeLLERTZ, S. fibrinolytic activity. Arch. Biochem. Biophys. 40, 346-351, 1952. 20.
NIEUWENHUIZEN, W., EMEIS, J.J., VERMOND, A., KURVER, P. and VAN DER HEIDE, D. Studies on the catabolism and distribution of fibrinogen in rats. Application of the IODOGEN labelling technique. Biochem. Biophys. Res. Commun. 97, 49-55, 1980.
21.
BLOMBACK, B., CARLSON, L.A., FRANZEN, S. and ZETTERQVIST, E. Turnover of 1311-labelled fibrinogen in man. ; Studies in normal subjects, in congenital coagulation factor deficiency states, in liver cirrhosis, in polycythemia vera and in epidermolysis bullosa. Acta. Med. Stand. 179, 557-574, 1966.
22.
DAY, J.F., THORNBURG, R.W., THORPE, S.R. and BAYNES, J.W. Nonenzymatic glucosylation of rat albumin. ; Studies in vitro and in vivo. J. Biol. Chem. 254, 9394-9400, 1979.
Purification of ra,t 23. VAN RUIJVEN-VERMEER, I.A.M. and NIEUWENHUIZEN, W. fibrinogen and its constituent chains. Biochem. J. 169, 653-658, 1978. Production of 24. MCVERRY, B.A., HOPP, A., FISHER, C. and HUEHNS, E.R. pseudodiabetic renal glomerular changes in mice after repeated injections of glucosylated proteins. Lancet 1, 738-740, 1980. Glucosylated 25. JERAJ, K.P., MICHAEL, A.F., MAUER, S.M. and BROWN, D.M. and normal human or rat albumin do not bind to renal basement membranes of diabetic and control rats. Diabetes 32, 380-382, 1983. 26. BROWNLEE, M.,
Inhibition of heparinVLASSARA, H. and CERAMI, A. catalyzed human antithrombin III activity by nonenzymatic glycosylation. Possible role in fibrin deposition in diabetes. Diabetes 33L 532-535, 1984.
27. VILLANUEVA, G.B. and
ALLEN, N. Demonstration of altered anti-thrombin III activity due to nonenzymatic glycosylation at glucose concentration Diabetes 37, expected to be encountered in severely diabetic patients. 1103-1107, 1988.
28.
GEIGER, M. and BINDER, B.R. Plasminogen activation in diabetes mellitus Kinetic of plasmin formation with tissue plasminogen activator and plasminogen from individual diabetic donors and with in vitro glucoEnzyme 40L 149-157, 1988. sylated plasminogen.
INCREASED ACCUMULATION OF NONENZYMATICALLY GLYCATED* FIBRINOGEN IN THE RENAL CORTEX IN RATS.
Takashi Murakami, Hiroshi Egawa, Yutaka komiyama, Midori Masuda, and Kenjiro Murata Department of Clinico-laboratory Medicine, Kansai Medical University, Moriguchi, Osaka 570, JAPAN (Received 19.10.1989; accepted in revised form 16.1 .1990 by Editor S. Okamoto)
ABSTRACT We have determined that the nonenzymatic glycation of fibrinogen altered its biological functions in vitro. Thrombin clottability of rat fibrinogen incubated with glucose decreased with increasing incubation time, but was not affected by the glucose concentration. Fibrin prepared from glycated fibrinogen showed a significant resistance in susceptibility to plasmin degradation. We also examined the in vivo distribution of glycated fibrinogen in renal cortex. Iodine labeled rat glycated or unglycated fibrinogen was injected into streptozocin-induced diabetic and control rats. NO appreciable difference in the plasma disappearance rate in control rats was observed (half-lives in hours for glycated, 25.6 f 0.37; unglycated, 26.1 k 0.74). The radioactivity of fibrinogen retained to the renal cortex was calculated 2,!+-hours after injection. In the retention rate of glycated both control and diabetic rats, fibrinogen in renal cortex was significantly higher than that of the unglycated. These results suggest that glycated fibrinogen may fibrin occur in a more resistant form to plasmin digestion with deposition as confirmed in in vitro studies. Therefore, we suggest that glycated fibrinogen may partly contribute to the development of diabetic microangiopathic lesions such as glomerulosclerosis.
Key words : Nonenzymatic glycation (Glycosylation), Rat fibrinogen, Diabetic microangiopathies,
Fibrin deposition.
*The Joint Commission on Biochemical Nomenclature of the International Union of Biochemistry (IUB) and the International Union of Pure and Applied Chemistry (IUPAC) suggests the term glycation rather than glycosylation (or glucosylation) for the nonenzymatic reaction that links a glucose or other sugars to a protein (6). 23
24
RAT GLYCATED FIBRINOGEN IN VIVO
Vol. 58, No. 1
INTRODUCTION Fibrin(ogen) is one of several proteins that have been immunohistochemically demonstrated in diabetic angiopathic lesions, especially in renal glomerular basement membranes (GBM) and mesangium (1). In the glomerus, the deposition of these proteins causes increasing permeability and expanding Finally mesangium, and leads to secondary accumulation of proteins. capillary occlusion and nodular droplets occur. On the other hand, in diabetic patients, various abnormalities in coagulation and fibrinolysis are found (2-5), i.e., higher fibrinogen concentrations, elongation of the thrombin clotting time, reduced fibrinogen survival, and elevated fibrinopeptide A and fibrin degradation products. It is not known whether these abnormalities are due to fibrinogen itself. Because some of these abnormalities are reversed by appropriate treatment for hyperglycemia (3,4), there is a possibility that fibrinogen alters its biological function during hyperglycemia. Recently it has been reported that nonenzymatic glycation (6) of fibrinogen increased in poorly controlled diabetic patients as well as other proteins (7-IO), and that glycated fibrinogen modified its native biological functions in vitro (11-14). Therefore, we hypothesized that fibrin deposition as seen in diabetic microangiopathic lesions is due to increased glycated fibrinogen which alters its native biological functions. In the present studies, we injected labeled glycated fibrinogen into streptozocin (STZ) induced diabetic and control rats. We compared plasma half-life and renal cortical accumulation of glycated fibrinogen with those of the unglycated.
MATERIALS AND METHODS Animals Male Sprague-Dawley rats, 9 weeks old and weighing 230-250 g, were purchased from Japan SLC. (Hamamatsu, Japan). Diabetes was induced by intravenous injection of 25 mg/kg STZ (lot 605527, Calbiochem, La Jolla, CA), which was freshly dissolved in 0.9% saline acidified with 10 mM citric acid pH 4.2. Diabetes was confirmed by the presence of hyperglycemia, which was checked once a week for 3 months after the administration of STZ. Animals with nonfasting plasma glucose levels of less than 360 mg/dl were rejected. Diabetic rats and age-matched control rats were allowed free access to food and water. In vivo experiments were performed after 3 months of maintenance. Glycation of fibrinogen Fibrinogen from fresh rat plasma was purified by cold ethanol precipitation (15), and contaminating plasminogen was removed by chromatography on lysine-Sepharose (Pharmacia, Uppsala, Sweden) by the method of Matsuda et a1.(16). This purified fibrinogen (92% clottable) was dissolved in 50 mM sodium phosphate buffer pH 7.4 with 0.15 M sodium chloride (PBS), and glucose added to this. The solutions were sterilized by passing through a 0.45 urnMillipore filter and then incubated for up to 96 hours at 37°C. Free glucose was removed by dialysis extensively at 4°C against PBS. Quantification of glycated fibrinogen was determined by affinity chromatography by means of a m-aminophenylboronic acid-bound column
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN VIVO
25
(Glyco-Gel B; Pierce, Rockford, IL) as previously described (IO, 17). Briefly, fibrinogen solution was dialyzed against 0.25 M sodium ammonium buffer pH 8.5, and then applied to a column which was pre-equilibrated with the same buffer. After collecting the unglycated fraction passed through, the bound glycated fibrinogen was eluted with 0.25 M sodium acetate buffer The percentage of glycation (%glycation) was measured by pH 5.5. calculating the total absorbance at 280 nm (Aneo) of each fibrinogen fraction. Biological function of the glycated fibrinogen in vitro Thrombin clottability of incubated fibrinogen was measured by the method of Laki (18). For estimating the degradation of fibrin formed from glycated fibrinogen (glycated fibrin) by plasmin digestion, a modified fibrin plate assay (19) was performed. Briefly, the glycated (incubated with 500 mM glucose for 72 hours) and unglycated (without glucose, 72 hours) fibrinogen was made by dissolving l+mg/ml in PBS. Agarose (Agarose L; low gel point type, Wako Pure Chemicals, Osaka, Japan) was dissolved in PBS containing 0.02% sodium azide. Three ml of fibrinogen solution and 7 ml of agarose solution were mixed in each plastic plate at /+2"C, and allowed to cool to room temperature. After setting to gel, the plates were clotted by adding 3 ml of 10 U/ml of thrombin and incubated at 37°C for 2 hours. Exactly 3 ~1 of different concentrations (1, 6 casein units (ClJ)/ml)of standard plasmin solution (KabiVitrum, Stockholm, Sweden) was placed in a The well which was holed in the surface of the fibrin plates. by plasmin was determined by susceptibility of fibrin to degradation measuring the mean of two perpendicular diameters of the digested fibrin area. Iodine labelling The glycated fibrinogen used in the following experiments was made by incubation with 500 mM glucose for 72 hours, and was separated from unglycated fractions by borate affinity chromatography. Glycated and unglycated fibrinogen were iodinated using Na1251 (Du Pont Company, Wilmington, DE) and 1,3,4,6-tetrachloro-3a,6a-diphenylglycouril(IODOGEN ; Pierce) by the method of Nieuwenhuizen (20). Assessment of clottable radioactivity in the labeled fibrinogen was performed according to Blomback et al. (21). The final specific radioactivities and clottable activities were 6.6 x 10' cpmlmg, 86.2% of glycated, and 6.7 x IO7 cpm/mg, 88.0% of unglycated While iodination was carried out after glycation, it was fibrinogen. possible that the conditions of iodination might modify the nature of the glucosyl adduct (22), so we checked whether the glycated fibrinogen still had affinity to the boronate-Sepharose column after iodination (data not shown). These were stored at -70°C until use. The circulating half-lives of labeled fibrinogens Control rats (10 animals, weighing 240 + 7.5 g) were injected via the jugular vein with either 0.06 mg of 1251-glycated or '251-unglycated fibrinogen under mild ether anesthesia. A series of blood samples (0.4 ml each) was drawn from the opposite jugular vein with a heparinized syringe at 6 minutes and over a 6-day period. Radioactivity was determined in 0.1 ml aliquots of plasma precipitated with an equal volume of 20% trichloroacetic acid (TCA-precipitated plasma).
26
RAT GLYCATED FIBRINOGEN IN VW0
Vol. 58, No. 1
Accumulation of glycated fibrinogen in renal cortex in vivo Labeled fibrinogens were diluted with saline and final concentrations were 0.2& mg/ml. Control and diabetic rats were divided randomly into two subgroups. These animals were injected with either glycated or unglycated fibrinogen (1 ml per 1 kg rat body weight), and blood samples were drawn at 6 minutes. At 2f,-hoursafter injection, to remove the unbound fibrinogen Rats were from renal cortex, rat kidneys were perfused as follows. anesthetized with ether and the abdominal aorta exposed through an abdominal incision. The abdominal aorta and inferior vena cava were ligated just above the right renal veins, and kidneys were then perfused via the abdominal aorta and the renal arteries with 100 ml of saline containing 5 U/ml heparin sulfate. Kidneys were extirpated and weighed, then the medullae were excised. The radioactivity of these cortices was measured directly and expressed as counts per minute (cpm) per wet weight of renal cortices (g). The retention rate of labeled fibrinogen in renal cortex was given as percent of radioactivity in TCA-precipitated plasma at 6 minutes. Analytical procedures Blood glucose levels, blood urea nitrogen and serum creatinine concentrations were measured by an autoanalyzer (Beckman, Brea, CA). Plasma fibrinogen concentrations were measured by the thrombin time method. Protein concentration was determined by measuring Aneo (Al%/lcm 280nm = 15.4 (23) for rat fibrinogen). The radioactivity of samples was measured in a gamma spectrometer (Packard, Downers Grove, IL). Statistics Results are expressed as means + SE. Statistical methods used were Student's t-test for paired observations and linear-regression analysis with t-test of regression coefficient. P values reported were for one-tailed tests and values < 0.05 were considered significant.
RESULTS In
vitro
experiments
As shown in Fig. 1, the %glycation of fibrinogen was dependent on the length of incubation time and glucose concentration, but in 500 mM glucose it reached a plateau at 72 hours. The thrombin clottability of these fibrinogen decreased with time, but was not affected by the glucose concentration (Fig. 2). We also examined sodium dodecylsulfate polyacrylamide gel electrophoresis to determine the molecular degradation of each incubated fibrinogen, it showed identical patterns (data not shown). These results indicate that glucose, which covalently binds to the fibrinogen molecule, did not modify the thrombin clotting mechanism. Fibrin plate assay indicated the time course of plasmin digestion for glycated fibrin compared with that from the unglycated (Fig. 3), and it revealed that glycated fibrin was resistant to degradation by plasmin.
RAT GLYCATED FIBRINOGEN IN WV0
Vol. 58, No. 1
INCUBATION
TIME
27
( H )
FIG. 1. The %glycation of fibrinogen determined by using boronate affinity chromatography. Rat fibrinogen, 2.5 mg/ml of PBS, was incubated with each 20 mM (A), 50 mM (a), 100 mM (A,), 500 mM (0) glucose at 37°C. At indicated times the incubation mixture was dialyzed against 0.25 M ammonium acetate pH 8.5 at 4°C overnight, and 2 ml of this solution was applied to the column (bed volume: 4 ml). Each point represents the mean f SE of 3 separate experiments performed in duplicate.
100
X -
1
1
’
0
I
20 GLUCOSE
50 CONCENTRATION
100 (
500
mM 1
FIG. 2. Time course of thrombin clottability of rat fibrinogen incubated Fibrinogen, 2.5 mg/ml of PBS, was incubated with each with glucose. indicated concentration of glucose at 37°C for 24 (01, 48 (A,, 72 (A), 96 (0) hours. Each point represents the mean f SE.
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN WV0
12.5 -
. 6
12 INCUBATION
24 TIME
( H 1
FIG. 3. Effects of plasmin digestion with fibrin plates made from glycated and unglycated fibrinogen. Fibrinogen, 10 mg/ml of PBS, was incubated with or without 500 mM glucose at 37°C for 72 hours. Fibrin plates were prepared from glycated (0) and unglycated (0) fibrinogen, and were digested with 6 CU/ml (closed lines) and 1 CU/ml (broken lines) plasmin. Rate of digestion by plasmin was determined by measuring the mean of two perpendicular diameters of the digested fibrin area. Each point represents the mean + SE of 25 determinations. *P < 0.01 vs. unglycated.
In vivo
experiments
Laboratory and physiological data are shown in Table 1. Despite the loss of body weight in diabetic rats, the kidney weights were significantly greater than those of control rats. The plasma fibrinogen concentrations were significantly reduced in diabetic rats compared with control rats. The creatinine values were significantly increased in diabetic rats.
TABLE
1.
Characteristics of Experimental Animals at Time of Sacrifice Control (n = 20) Body weight (g) Kidney weight (g) Blood glucose (mg/dl) Fibrinogen (mg/dl) Blood urea nitrogen (mg/dl) Serum creatinine (mg/dl) Values are means f SE.
449.6 2.83 110.3 260.2 20.9 0.51
* + f * * *
3.0 0.28 4.5 6.2 0.46 0.018
N.S. = not significant.
Diabetic (n = 18) 307.5 3.08 518.1 185.8 22.2 0.68
f + * + f f
8.5 0.09 13.7 9.8 0.92 0.092
*P < < < <
0.01 0.05 0.01 0.01 N.S. < 0.05
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN WV0
29
In vivo experiments showing the kinetics of clearance of labeled fibrinogen were performed (Fig. 4). These radioactivity data between 24 and 124 hours possessed the linearity, and no significant difference in plasma disappearance rate for fibrinogen was found in the control rat circulation. The plasma radioactivity half-lives of glycated and unglycated fibrinogen were 25.6 * 0.4 hours and 26.1 + 0.7 hours, respectively. This result documents that glycation process does not affect the biological half-life of fibrinogen.
Fig. 5 shows the accumulation of labeled glycated and unglycated fibrinogen in renal cortex 24-hours after injection. In both control and diabetic rats, the retention rate of glycated fibrinogen in renal cortex was significantly higher than that of the unglycated. But each glycated and unglycated fibrinogen had lower retention rate in diabetic rats compared with that in control rats.
100 50
x -
10
c ; : s d g
1
f x
0.1
I
I
I
I
I
24
40
72
96
120
TIME
( H 1
Time course of disappearance of '251-glycated and unglycated FIG. 4. fibrinogen from the rat circulation. Control animals (five per experiment) were intravenously injected with either 0.06 mg of '251-glycated (0) or The disappearance of these labeled '*'I-unglycated (0) fibrinogen. fibrinogen from the circulation monitored in blood samples taken at Plasma fibrinogen radioactivities were expressed as indicated times. percentage of initial value, which was based on radioactivity at 6 minutes. Each point represents the mean f SE.
RAT GLYCATED FIBRINOGEN IN WV0
t
i-7
_I m
Vol. 58, No. 1
Glycated Unglycated
0
CONTROL
DIABETIC
Accumulation of '*'I-glycated and unglycated fibrinogen in renal FIG. 5. cortex. The retention rate was determined as described under MATERIALS AND METHODS. Each value represents the mean + SE. *P < 0.01 vs. unglycated.
DISCUSSION To prove the hypothesis that the glycated fibrinogen play an important role in the development of diabetic angiopathy, we have attempted to experimental demonstrate glycated fibrinogen in microvessels using caused animals. Repeated injections of glycated plasma proteins in mice GBM thickening and mesangial changes similar to those observed by electron microscopy in humans and experimentally induced diabetes (24). On the other hand, exogenous glycated albumin does not bind to GBM in diabetic or control However, these rats by means of immunofluorescence techniques (25). analytical methods make no distinction between injected proteins and endogenous protein deposition. If we can directly demonstrate in diabetic kidneys that the proportion of glycated fibrin(ogen) which accumulates in CBM and mesangial areas is greater than that which is present in plasma, it would enable us to prove that glycated fibrinogen contributes to the development of microangiopathic complications. But nonenzymatic glycated fibrin(ogen) is identified with unglycated fibrinogen by conventional immunohistochemical methods. So in the present studies, we used radiolabeled materials to reveal that glycated fibrinogen attached to renal cortex at higher concentrations than the unglycated. We demonstrated that the half-lives of radiolabeled glycated and unglycated fibrinogen in vivo were about l-day, which was compatible with results reported by Nieuwenhuizen et a1.(20). Glycation did not affect the biological half-life of fibrinogen as it did that of albumin (23). Since the clottabilities of these fibrinogens were equal, it could be speculated that the glycation process did not denature the fibrinogen molecule, and
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN WV0
31
also did not increase fibrinogen consumption. It further confirms the results of Jones et a.Z.(3)who reported that reduced fibrinogen survival in diabetes patients was not due to modified fibrinogen molecules. As the fibrin plate assay revealed that glycated fibrin was resistant to degradation by plasmin, glycated fibrinogen may be in a more insoluble form. This result confirmed a previous report of Brownlee et a1.(12). On the contrary, Ney et a1.(13) reported that the degradation of fibrin clots formed from glycated and unglycated fibrinogen were identical, but they used material which was incubated with a 20 mM glucose concentration. Our data using affinity chromatography indicated that the Xglycation under such conditions was maintained at a low level. It seems likely that differences in degradation rates were not observed because of little amounts of glycated fibrinogen in their materials. In both diabetic and control rats, we showed that the retention rate of glycated fibrinogen to renal cortex was still higher than that of the unglycated. But each glycated and unglycated fibrinogen had lower retention Although this rate in diabetic rats compared with that in control rats. reason is not clear, we suggest that some coagulation and fibrinolytic disorders which existed in diabetic states might interfere the fibrin deposition (26-28) or that the proliferation of endogenous fibrinogen binding to renal GBM during 3-months of STZ induced diabetes prevented the attachment of administered fibrinogen. In any case, we concluded that enhanced accumulation of glycated fibrin(ogen) in renal cortex was caused by the its resistance to degradation, and thus it seems to be evidence of initial phase of diabetic microangiopathic changes. Some investigators have assumed that glycated fibrinogen is physiologically not important because of the short half life of fibrinogen (humans, 3 - 4 days; rats, 1 day ) and the plasma glucose levels as low as 50 mM even in severe diabetic patients resulting in the glycated fibrinogen to appear in a small proportion in the circulation (13, 14). However, from our results, it seems that in chronic hyperglycemic states, a continuous exposure to glycated fibrinogen in the microcirculation might not be underestimated. Therefore, we suggest that glycated fibrinogen may partly contribute to the development of diabetic microangiopathic lesions such as glomerulosclerosis.
REFERENCES Diabetes mellitus and gout. In: Pathology of the 1. HEPTINSTALL, R.H. Kidney, 3rd edition. Little, Brown and Co., Boston, pp.139'7-1453,1983. Fibrinocoagulopathy in matu2. BANERJEE, R.N., SAHNI, A.L. and KHMAR, V. Thromb. Diath. and atherosclerosis. rity onset diabetes mellitus Haemorrh. 30, 123-132, 1973. Reduced fibrinogen survival in diabetes 3. JONES, R.L. and PETERSON, C.M. mellitus. A reversible phenomenon. J. Clin. Invest. 63, 485-493, 1979. Fibrinopeptide-A in diabetes mellitus. Relation to levels 4. JONES, R.L. of blood glucose, fibrinogen disappearance, and hemodynamic changes. Diabetes 34, 836-843, 1985.
32
RAT GLYCATED FIBRINOGEN IN WV0
Vol. 58, No. 1
Lower molecular weight fibrinogen and 5. GUREWICH, V. and LIPINSKA, I. non-clottable fibrinogen derivatives in diabetic mellitus. Thromb. Res. 22, 535-541, 1981. 6. NOMENCLATURE COMMITTEE OF IUB (NC-IuB) : IUB-IUPAC JOINT COMMISSION ON BIOCHEMICAL NOMENCLATURE (JCBN). Hoppe-Seyler's Z. Physiol. Chem. 365, I-V, 1984. Glycosyla7. LCTJENS, A., TE VELDE, A.A., VD VEEN, E.A. and VD MEER, J. Diabetologia 28, 87-89, 1985. tion of human fibrinogen in vivo. 8. KOENIG, R.J., PETERSON, C.M., JONES, R.L., SAUDEK, C., LEHMAN, M, and CERAMI, A. Correlation of glucose regulation and hemoglobin Ale in diabetes mellitus. N. Engl. J. Med. 295, 417-420, 1976. Nonenzymatically glucosylated 9. DAY, J.F., THORPE, S.R. and BAYNES, J.W. albumin. ; In vitro preparation and isolation from normal human serum. J. Biol. Chem. 254, 595-597, 1979. 10. KANESHIGE, H. Nonenzymatic glycosylation of serum IgG and its effect on antibody activity in patients with diabetes mellitus. Diabetes 36, 822828, 1987. 11. MCVERRY, B.A., THORPE, S., JOE, F., GAFFNEY, P. and HUEHNS, E.R. Nonenzymatic glucosylation of fibrinogen. Haemostasis 10, 261-270, 1981. 12. BROWNLEE, M., VLASSARA, H. and CERAMI, A. Nonenzymatic glycosylation reduces the susceptibility of fibrin to degradation by plasmin. Diabetes 32, 680-684, 1983. 13. NEY, K.A., PASQUA, J.J., COLLEY, K.J., GUTHROW, C.E. and PIZZO, S.V. In vitro preparation of nonenzymatically glucosylated human transferrin, a2-macroglobulin, and fibrinogen with preservation of function. Diabetes 34, 462-470, 1985. 14. MIRSHAHI, M., SORIA, J., SORIA, C., BERTRAND, O., MIRSHAHI, M. and BASDEVANT, A. Glycosylation of human fibrinogen and fibrin in vitro its consequences on the properties of fibrin(ogen). Thromb. Res. 48, 279-289, 1987. 15. DOOLITTLE, R.F., SCHUBERT, D. and SCHWARTZ, S.A. Amino acid sequence studies on artinodactyl fibrino-peptides. Biochem. Biophys. Acta. 118, 456-467, 1967. 16. MATSUDA, M., IWANAGA, S. and NAKAMURA, S. A simple, large scale method for preparation of plasminogen-free fibrinogen. Thromb. Res. 1, 619630, 1972. 17. MALLIA, A.K., HERMANSON, G.T., KROHN, R.I., FUJIMOTO, E.K. and SMITH, P.K. Preparation and use of a boronic acid affinity support for separation and quantitation of glycosylated hemoglobins. Anal. Lett. 14, 649-661, 1981. 18. LAKI, K. The polymerization of proteins. ; The action of thrombin on fibrinogen. Arch. Biochem. Biophys. 32, 317-324, 1951.
Vol. 58, No. 1
RAT GLYCATED FIBRINOGEN IN VIVO
33
The fibrin plate method for estimating 19. ASTRUP, T. and MeLLERTZ, S. fibrinolytic activity. Arch. Biochem. Biophys. 40, 346-351, 1952. 20.
NIEUWENHUIZEN, W., EMEIS, J.J., VERMOND, A., KURVER, P. and VAN DER HEIDE, D. Studies on the catabolism and distribution of fibrinogen in rats. Application of the IODOGEN labelling technique. Biochem. Biophys. Res. Commun. 97, 49-55, 1980.
21.
BLOMBACK, B., CARLSON, L.A., FRANZEN, S. and ZETTERQVIST, E. Turnover of 1311-labelled fibrinogen in man. ; Studies in normal subjects, in congenital coagulation factor deficiency states, in liver cirrhosis, in polycythemia vera and in epidermolysis bullosa. Acta. Med. Stand. 179, 557-574, 1966.
22.
DAY, J.F., THORNBURG, R.W., THORPE, S.R. and BAYNES, J.W. Nonenzymatic glucosylation of rat albumin. ; Studies in vitro and in vivo. J. Biol. Chem. 254, 9394-9400, 1979.
Purification of ra,t 23. VAN RUIJVEN-VERMEER, I.A.M. and NIEUWENHUIZEN, W. fibrinogen and its constituent chains. Biochem. J. 169, 653-658, 1978. Production of 24. MCVERRY, B.A., HOPP, A., FISHER, C. and HUEHNS, E.R. pseudodiabetic renal glomerular changes in mice after repeated injections of glucosylated proteins. Lancet 1, 738-740, 1980. Glucosylated 25. JERAJ, K.P., MICHAEL, A.F., MAUER, S.M. and BROWN, D.M. and normal human or rat albumin do not bind to renal basement membranes of diabetic and control rats. Diabetes 32, 380-382, 1983. 26. BROWNLEE, M.,
Inhibition of heparinVLASSARA, H. and CERAMI, A. catalyzed human antithrombin III activity by nonenzymatic glycosylation. Possible role in fibrin deposition in diabetes. Diabetes 33L 532-535, 1984.
27. VILLANUEVA, G.B. and
ALLEN, N. Demonstration of altered anti-thrombin III activity due to nonenzymatic glycosylation at glucose concentration Diabetes 37, expected to be encountered in severely diabetic patients. 1103-1107, 1988.
28.
GEIGER, M. and BINDER, B.R. Plasminogen activation in diabetes mellitus Kinetic of plasmin formation with tissue plasminogen activator and plasminogen from individual diabetic donors and with in vitro glucoEnzyme 40L 149-157, 1988. sylated plasminogen.