- Email: [email protected]
Intervention strategies to prevent pathogenetic effects of glycated albumin
ABB Archives of Biochemistry and Biophysics 419 (2003) 25–30 www.elsevier.com/locate/yabbi
Minireview
Intervention strategies to prevent pathogenetic effects of glycated albumin Margo P. Cohen* Institute for Metabolic Research, University City Science Center, 3508 Market Street, Suite 420, Philadelphia, PA 19104, USA Received 25 January 2003, and in revised form 13 July 2003
Abstract Modification of proteins by nonenzymatic glycation is one of the underlying factors contributory to the development of complications of diabetes. In general, the nature of this structural modification falls into two broad categories: nonenzymatic glycation per se, which refers to the attachment of free carbohydrate to proteins in the Amadori construct, and Advanced Glycation Endproducts (AGE), which refers to a heterogeneous group of carbohydrate-modified products generated from the Amadori adduct by oxidation, polymerization, and other spontaneous reactions. This review will focus on the role of nonenzymatically glycated proteins, and in particular glycated serum albumin, in the pathogenesis of diabetic complications, and on pharmacologic approaches to mitigate their deleterious effects. Potential intervention strategies to lessen the influence of AGE-modified proteins, as well as of other contributory abnormalities, are discussed elsewhere in this volume. 2003 Elsevier Inc. All rights reserved.
The pathogenesis of the long-term complications of diabetes, which encompass microvascular, macrovascular, and neuropathic diseases, likely derives from diverse perturbations in cell biology superimposed on a polygenic predisposition and acting in concert with hyperglycemia, the defining metabolic abnormality of diabetes. Among the identified contributory processes are hemodynamic and oxidative stress, activation of various enzymatic pathways involved with glucose metabolism, stimulation of growth factor systems, activation of cell signaling pathways, modulation of responsivity to hormonal, growth factor, and cytokine influences, and acceleration of the nonenzymatic glycation of circulating and tissue proteins (Table 1). Understanding the role of and mechanisms by which these elements participate in the development of diabetic complications has advanced substantially in recent years. This review will focus on evidence linking increased nonenzymatic glycation of serum proteins to the pathophysiology of diabetic vasculopathies and potential avenues of therapeutic intervention to countermand the consequent deleterious effects. * Fax: 1-215-222-5325. E-mail address: [email protected].
0003-9861/$ - see front matter 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2003.08.012
Nonenzymatic glycation Nonenzymatic glycation is a condensation reaction between carbohydrate and free amino groups at the NH2 -terminus or susceptible e-amino groups of lysine residues of proteins. The reaction is initiated with attachment of the aldehyde function of acyclic glucose to the protein amino group via nucleophilic addition, forming an aldimine, also known as a SchiffÕs base. This intermediate product subsequently undergoes an Amadori rearrangement to form a 1-amino-deoxyfructose derivative in stable ketoamine linkage (Fig. 1), which in turn can cyclize to a ring structure [1–5]. The amount of the labile SchiffÕs base intermediate increases rapidly within a few hours after incubation of protein with glucose and reaches equilibrium [3], during which time the formation of the stable ketoamine is negligible. The rate of Amadori rearrangement has been calculated to be about 1/60th of the rate of dissociation of the aldimine back to glucose and protein, indicating that the rearrangement step is rate limiting for the formation of stable product. The glycation reaction follows secondorder kinetics and glucose is a major determinant of the amount of Amadori product formed. Another main determinant is the time of exposure of the protein to
26
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
Table 1 Pathogenetic factors in diabetic complications • • • • • • •
Polygenic predisposition Hemodynamic stress Oxidative stress Enzymatic shunting in glucose metabolism (polyol pathway; GFAT) Signal pathway activation (MAPK cascade; JAK/STAT) Stimulation of growth factor/cytokine systems (TGF-b; Angiotensin II: Thromboxane; Endothelins; IGF-I; PDGF: VEGF: Connective Tissue Growth Factor) Nonenzymatic glycation
Fig. 1. Condensation reaction between glucose and protein epsilon amino groups.
increased glucose concentration since glucoadducts will continue to form as a function of time until equilibrium is reached. In vivo, these two influences (glucose concentration and time) translate to degree and duration of hyperglycemia, and result in the slow and irreversible process of glycation during the life of the protein. Glycated albumin accounts for about 80% of the circulating glycated protein and albumin modified by Amadori glucose adducts is the predominant form in which glycated albumin exists in vivo. The principal residues subjected to glycation in vivo are lys-525, lys439, lys-281, and lys-199 [6,7]. Hyperglycemia attendant to the diabetic state increases nonenzymatically glycated albumin and the extent of increase reflects the time-averaged concentration of glucose to which the protein has been exposed during its residence time in the circulation. Like the measurement of the amount of glycated hemoglobin or HbA1c in red cells, determination of the serum concentration of glycated albumin assesses integrated glycemia over the relevant retrospective period, which is about 2–3 weeks in humans [4–6]. However, HbA1c serves only as a marker of glycemic status whereas glycated albumin has documented biologic effects, which have been causally linked to the pathogenesis of diabetic nephropathy and other complications. Moreover, as discussed below, these effects are
operative independent of hyperglycemia, indicating that they can continue to exert a pathogenetic influence in vivo during the half-life of the protein, even if normoglycemia is restored.
Biologic effects of glycated albumin The notion that exposure of the capillary beds to serum containing increased concentrations of glycated albumin due to diabetes could be deleterious to vascular physiology was suggested by early studies showing that transfusion into normal rats of Amadori-modified serum proteins in concentrations found in streptozotocin-diabetic rats induced hyperfiltration, a functional abnormality that has been implicated in the development of diabetic nephropathy [8]. Other earlier work in support of this hypothesis had indicated that glycated albumin is preferentially transported across the glomerular filtration barrier [9–11], that glomerular mesangial and epithelial cells exhibit an enhanced uptake of glycated albumin that is accompanied by an increase in cell hydrogen peroxide production [12,13], that nondiabetic mice injected with glycated plasma proteins develop glomerular basement membrane thickening [14], and that there is widespread connective tissue accumu-
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
27
Table 2 Comparability of biologic effects of glycated albumin and high glucose on glomerular cells in culture
Expression/Production of ECM proteins MAPK cascade Protein kinase C TGF-b system a b
Glycated albumina
Glucoseb
Stimulates Activates Activates Upregulates
Stimulates Activates Activates Upregulates
5.5 mM glucose media. P 25 mM glucose media.
lation of glycated albumin in diabetes [15–17]. More recent studies have uncovered a myriad of pathobiologic events induced by glycated albumin, which include increased gene expression of interleukin (IL)-8 and nitric oxide synthase and nitric oxide production in endothelial cells [18,19], activation of Protein Kinase C (PKC)1, the Mitogen Activated Protein Kinase (MAPK) cascade, and the Nuclear Factor (NF)-jB and increased gene expression of the inflammatory cytokines Monocyte Chemoattractant Peptide (MCP-1) and IL-6 in vascular smooth muscle cells [20,21], increased oxidative stress, phosphorylation of Extracellular Signal Regulated Kinase (ERK), and nuclear translocation of NFjB in monocyte–macrophages [22], and activation of signaling pathways and upregulation of growth factor systems in renal glomerular cells. Data from experiments with glomerular mesangial and endothelial cells in culture have established that albumin modified by Amadori glucose adducts induces significant alterations in glomerular cell biology that are highly reminiscent of abnormalities known to be associated with diabetic nephropathy and that resemble those induced by high glucose concentrations. Specifically, glycated albumin has been shown to: (a) stimulate the expression of a1 (IV) collagen and fibronectin, the predominant constituents of the expanded ECM that is seen in diabetes [23–28]; (b) activate PKC-b1 and ERK [29–31]; and (c) increase expression of mRNA encoding the fibrogenic Transforming Growth Factor (TGF)-b1 and its primary signaling receptor, the TGF-b type II receptor [27,31,32]. The modulations in cell biology induced by glycated albumin are seen with concentrations of glycated albumin representative of those found in clinical specimens and are analogous to but operate independent of those observed in response to high glucose concentration (Table 2), since they can be demonstrated in a physiologic (5.5 mM) glucose milieu. These in vitro data are consistent with the hypothesis that glycated albumin is a sufficient stimulus to set into motion a pathogenic program characterized by alterations in key cellular mediators that affect cell signaling pathways, 1 Abbreviations used: PKC, protein kinase C; MAPK, mitogen activated protein kinase; NF-jB, nuclear factor-jB; MCP-1, monocyte chemoattractant peptide; TGF-b1, transforming growth factor-b1; AGE, advanced glycation endoproducts.
which are important in regulating extracellular matrix (ECM) production. Together, the effects of glycated albumin and hyperglycemia may be additive, creating a self-reinforcing pathophysiologic cycle (Fig. 2). The hypothesis formulated from in vitro findings that glycated albumin is an independent and potent trigger of molecular mediators contributory to complications of diabetes is supported by results of in vivo studies showing that neutralizing its biologic effects [33–37] or inhibiting its formation [38–40] can ameliorate structural, cell biology, and functional abnormalities in the kidney and retinal microvasculature, and can do so in the face of persistent hyperglycemia.
Clinical considerations The traditional approach to forestalling the onset or progression of complications of diabetes has focused on reducing systemic risk factors and optimizing control of blood glucose, the potential benefits of the latter being underscored by the results of the Diabetes Control and Complications Trial (DCCT) and the more recent United Kingdom Prospective Diabetes Study (UKPDS) [41–44]. Despite wide publicity of the findings of these trials, which established that improved glycemic control reduces the risk for diabetic complications, the invigoration of efforts for bettering the glycemic status in diabetic patients, and the introduction of new anti-diabetic agents such as the thiazolidenediones, glycemic control remains woefully inadequate in the majority of people with diabetes. Normoglycemia is difficult to achieve and may be associated with hypoglycemic episodes of unacceptable frequency or severity [45]. Recent studies showing that modulation of hemodynamic influences with ACE inhibitors or angiotensin receptor (AR) antagonists can slow the inexorable progression of diabetic nephropathy offer promise [46–49], but these measures are neither curative nor completely effective. Thus, there is an appreciation that diabetic patients may require treatments specifically targeted at the vascular complications of the disease [50] and that, if possible, such agents should be effective, regardless of the glycemic status. Evidence from animal studies, discussed below, indicates that reduction of the burden of glycated albumin constitutes a viable therapeutic strategy to this
28
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
Fig. 2. Dual-pronged self-reinforcing pathogenetic processes triggered by glycated albumin and glucose. GA, glycated albumin, ECM, extracellular matrix.
end. Available clinical information also is supportive of this approach. In human diabetes, the concentration of Amadori-modified glycated albumin is independently associated with diabetic nephropathy and vascular dysfunction [51], and localization of Amadori-modified albumin in glomeruli of patients with diabetic nephropathy corresponds with the severity of renal involvement [52].
Therapeutic approaches Strong evidence supporting the view that glycated albumin is a potential target of therapy in the treatment of diabetic complications derives from studies in the db/ db mouse, a genetic model of diabetes that develops renal lesions resembling those observed in human diabetic nephropathy. The db/db mouse is usually considered a model of type 2 diabetes, but the early hyperglycemia, transition to an insulin deficient state, and development of classic renal pathology and functional abnormalities render it relevant for the study of diabetic nephropathy occurring in both type 1 and type 2 diabetes [53–59]. Diabetic nephropathy is best described in patients with type 1 diabetes, but similar structure–function changes are known to occur in patients with type 2 diabetes [60,61]. The first line of evidence comes from a series of studies examining the effects of treatment of db/db mice with a murine monoclonal antibody that specifically recognizes Amadori glucose adducts in rodent and human glycated albumin [62]. Chronic administration of the antibody was shown to reduce proteinuria and attenuate mesangial expansion, assessed by visual inspection, in diabetic db/db mice [33]. The reno-protective response to antibody treatment was interpreted to reflect a reduction in circulating biologically active glycated epitopes, since it was accompanied by a significant
decrease in the plasma glycated albumin concentration, ascribed to an enhanced clearance through the reticuloendothelial system. Subsequent experiments with rigorous glomerular morphometric analysis showed that treatment with the monoclonal antibody significantly reduced the accumulation of glomerular mesangial matrix, and that this reduction was accompanied by a decrease in the renal cortical overexpression of the extracellular matrix proteins fibronectin and type IV collagen, assessed by Northern blot [34]. Antibody treatment also prevented the thickening of the retinal microvessel basement membrane observed in control diabetic mice [37]. A similar program of chronic treatment with antibodies unreactive with glycated albumin was unable to duplicate these beneficial effects, corroborating the interpretation that neutralizing the biologic effects of excess Amadori-modified albumin in the circulation has a salutary influence on the development of renal and retinal pathology in diabetes. A second line of evidence comes from studies using small molecules that prevent the nonenzymatic glycation of albumin by binding to the protein at or near sites susceptible to glycation in vivo and rendering the involved e-amino groups inaccessible for condensation with glucose. The impetus to design such molecules was in part prompted by reports indicating that various antiinflammatory drugs could impede the in vitro attachment of free carbohydrate to protein [63–67]. New compounds have been synthesized that bind to wellcharacterized sites in serum albumin and that have potent anti-glycation properties and negligible cyclooxygenase inhibitory activity, and two of these have been shown to arrest the development of diabetic nephropathy in the db/db mouse [39,40]. After oral administration each of the compounds significantly lowered plasma concentrations of glycated albumin, confirming their ability to inhibit the condensation of glucose with glycatable sites in serum albumin in vivo.
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
They did not affect blood glucose concentrations, which would be an anticipated finding since the compounds do not react with free glucose and do not influence metabolic status. Despite persistent elevation of blood glucose concentrations, the drug-induced reductions in glycated albumin were accompanied by an amelioration of all of the functional and structural abnormalities characteristic of diabetic nephropathy, including significant decreases in the elevated urine protein excretion, the renal overexpression of ECM proteins, the mesangial matrix expansion, the falling creatinine clearance, and the rising serum creatinine. The in vivo inhibition of albumin glycation also lowered the elevated urinary excretion of type IV collagen, which reflects renal overproduction of this ECM protein and the associated glomerular matrix expansion that leads to glomerular occlusion and loss of filtration function in both human [68–77] and rodent [39,78,79] diabetes. Most notable is the finding that the treatment protocol significantly decreased the marked diabetes-associated glomerular overexpression of TGF-b1, assessed by in situ hybridization, thus closing the extrapolative loop between the in vitro observations that glycated albumin up-regulates the TGF-b system in glomerular mesangial and endothelial cells and the in vivo situation. The attenuation of the overexpression of glomerular TGF-b1 in conjunction with normalization of serum glycated albumin supports the hypothesis that increased glomerular TGFb1 expression in diabetes derives at least in part from the elevated circulating concentrations of glycated albumin associated with this disease. The data are consistent with the interpretation that the salutary effects on glomerulosclerosis (and, hence, compromised filtration function) associated with normalizing glycated albumin concentrations were mediated by a reduction in glomerular overexpression of TGF-b1, a potent fibrogenic cytokine that stimulates synthesis of ECM and inhibits its degradation and that has been strongly linked to diabetic glomerulosclerosis [80–84].
Conclusions and perspectives The studies from our laboratory and others that are recapitulated in this review have revealed that exposure of vascular beds to increased glycated serum proteins is an important contributor to the pathogenesis of diabetic complications. The evidence provided indicates that glycated albumin fulfills, to a large extent, KochÕs criteria for an etiologic agent in disease processes and that, once formed, its acquired biologic effects do not depend on a hyperglycemic milieu to be operative. Although hyperglycemia is the driving force for accelerated nonenzymatic glycation of proteins, it is clear from these studies that increased albumin glycation in diabetes can be therapeutically prevented in the presence
29
of hyperglycemia, and that therapeutic intervention to reduce the level of glycated albumin can arrest the evolution of diabetic nephropathy independent of antihyperglycemic therapy. The strategy of blocking the formation or effects of glycated albumin holds promise as a valuable therapeutic adjunct for the prevention and treatment of complications in human diabetes.
Acknowledgments This work was supported in part by research grants from the National Institutes of Health (DK54143; DK54608; and EY11825).
References [1] R.J. Koenig, S.H. Blobstein, A. Cerami, J. Biol. Chem. 256 (1981) 5204–5208. [2] H.F. Bunn, K.H. Gabbay, P.M. Gallop, Science 200 (1978) 21–27. [3] P.H. Higgins, H.F. Bunn, J. Biol. Chem. 256 (1981) 5204–5208. [4] J.F. Day, C.G. Ingelbretsen, W.R. Ingelbretsen, et al., Diabetes 29 (1980) 524–527. [5] C.E. Guthrow, M.A. Morris, J.F. Day, et al., Proc. Natl. Acad. Sci. USA 76 (1979) 4258–4261. [6] R.L. Garlick, J.S. Mazer, J. Biol. Chem. 258 (1983) 6142–6146. [7] N. Iberg, R. Fluckiger, J. Biol. Chem. 261 (1986) 13542–13546. [8] M. Sabbatini, G. Sansone, E. Ucello, et al., Kidney Int. 42 (1992) 875–881. [9] S.K. Williams, J.J. Devenny, M.W. Bitensky, Proc. Natl. Acad. Sci. USA 78 (1981) 2392–2397. [10] S.K. Williams, N.J. Solenski, Microvasc. Res. 28 (1984) 311–321. [11] S.K. Williams, R.K. Siegal, Kidney Int. 28 (1985) 146–152. [12] M. Agarwal, B.S. Daniels, Clin. Res. 40 (1992) 179A. [13] B.S. Daniels, R. Brandt, M. Agarwal, J. Am. Soc. Nephrol. 4 (1993) 791A. [14] B.A. McVerry, A. Hopp, C. Fisher, et al., Lancet 2 (1980) 738– 740. [15] K. Miller, A.F. Michael, Diabetes 25 (1976) 701–708. [16] A. Michael, D.M. Brown, Diabetes 30 (1981) 843–846. [17] S.B. Kelly, J.E. Olerud, J.L. Witztum, et al., J. Invest. Dermatol. 93 (1989) 327–331. [18] A. Amore, P. Cirina, S. Mitola, et al., Lab Invest. 51 (1997) 27–35. [19] Z.M. Bain, V.M. Elner, N.W. Lukacs, et al., Curr. Eye Res. 17 (1998) 65–72. [20] Y. Hattori, H. Kakishita, K. Akimoto, M. Matsumura, K. Kasai, Biochem. Biophys. Res. Commun. 281 (2001) 891–896. [21] Y. Hattori, M. Suzuki, S. Hattori, K. Kasai, Hypertension 39 (2002) 22–29. [22] M. Cohen, E. Shea, S. Chen, C.W. Shearman, J. Lab. Med. 141 (2003) 242–249. [23] M.P. Cohen, F.N. Ziyadeh, Kidney Int. 45 (1994) 475–484. [24] F.N. Ziyadeh, M.P. Cohen, Mol. Cell. Biochem. 125 (1993) 19–25. [25] M.P. Cohen, E. Hud, V.Y. Wu, F.N. Ziyadeh, Mol. Cell. Biochem. 151 (1995) 61–67. [26] M.P. Cohen, F.N. Ziyadeh, J. Am. Soc. Nephrol. 7 (1996) 183–190. [27] F.N. Ziyadeh, D.C. Han, J.A. Cohen, et al., Kidney Int. 53 (1998) 631–638. [28] M.P. Cohen, V.-Y. Wu, J.A. Cohen, Biochem. Biophys. Res. Commun. 139 (1997) 91–94.
30
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
[29] M. Cohen, F.N. Ziyadeh, G.T. Lautenslager, et al., Am. J. Physiol. 276 (1999) F684–F690. [30] M.P. Cohen, E.A. Shea, C.W. Shearman, Biochem. Biophys. Res. Commun. 283 (2001) 641–643. [31] S. Chen, M.P. Cohen, G.T. Lautenslager, et al., Kidney Int. 59 (2001) 673–681. [32] S. Chen, M.P. Cohen, F.N. Ziyadeh, Kidney Int. 58 (2000) 40–44. [33] M.P. Cohen, E. Hud, V.Y. Wu, Kidney Int. 45 (1994) 1673–1979. [34] M.P. Cohen, K. Sharma, K. Jin, et al., J. Clin. Invest. 95 (1995) 2338–2345. [35] M.P. Cohen, R.S. Clements, J.A. Cohen, C.W. Shearman, Diabetologia 39 (1996) 270–274. [36] M.P. Cohen, R.S. Clements, J.A. Cohen, C.W. Shearman, Biochem. Biophys. Res. Commun. 218 (1996) 72–75. [37] R.S. Clements, W.G. Robison, M.P. Cohen, J. Diabetic. Complic. 12 (1998) 28–33. [38] M.P. Cohen, N. Mason, E. Hud, R.S. Clements, Exper. Nephrol. 8 (2000) 135–143. [39] M.P. Cohen, E. Hud, E. Shea, C.W. Shearman, Metabolism 51 (2002) 901–905. [40] M.P. Cohen, F.N. Ziyadeh, S.W. Hong, C.W. Shearman, et al., Kidney Int. 61 (2002) 2025–2032. [41] The DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin dependent diabetes mellitus, N. Engl. J. Med. 329 (1993) 977–986. [42] The DCCT Research Group: Effect of intensive therapy on the development and progression of diabetic nephropathy in the Diabetes Control and Complications Trial, Kidney Int. 47 (1995) 1703–1720. [43] UK Prospective Diabetes Study Group: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes, Lancet 352 (1998) 837–853. [44] UK Prospective Diabetes Study Group: Effect of intensive bloodglucose control with metformin on complications in overweight patients with type 2 diabetes, Lancet 352 (1998) 854–865. [45] The DCCT Research Group: Epidemiology of severe hypoglycemia in the diabetes Control and Complications Trial, Am. J. Med. 90 (1991) 450–459. [46] E.J. Lewis, L.G. Hunsicker, et al., N. Engl. J. Med. 329 (1993) 1456–1462. [47] E.J. Lewis, L.G. Hunsicker, et al., N. Engl. J. Med. 345 (2001) 851–860. [48] B.M. Brenner, M.E. Cooper, et al., N. Engl. J. Med. 354 (2001) 861–869. [49] T.H. Hostetter, N. Engl. J. Med. 354 (2001) 910–911. [50] E. Duh, L.P. Aiello, Diabetes 48 (1999) 1899–1906. [51] C.G. Schalkwijk, N. Ligtvoet, H. Twaalfhoven, et al., Diabetes 48 (1999) 2446–2453. [52] H. Sakai, K. Jinde, D. Suzuki, et al., Nephrol. Dial. Transplant. 11 (Suppl. 5) (1996) 66–71. [53] K. Hummel, M.M. Dica, D.L. Coleman, Science 153 (1966) 1127–1128.
[54] D.L. Coleman, K.P. Hummel, Diabetologia 1 (1974) 607–610. [55] W.L. Chick, A. Like, Diabetologia 6 (1970) 243–251. [56] A.A. Like, R.L. Levine, P.L. Poffenbarger, W.L. Chick, Am. J. Pathol. 66 (1972) 193–204. [57] S.M. Lee, A. Graham, Exp. Mol. Pathol. 33 (1980) 323–332. [58] S.M. Lee, R. Bressler, Diabetes 30 (1981) 106–111. [59] M.P. Cohen, R.S. Clements, E. Hud, et al., Exper. Nephrol. 4 (1996) 166–171. [60] A. Ueno, S. Kawashima, S. Nihsi, et al., Kidney Int. 54 (1997) S191–S194. [61] K.E. White, R.W. Bilous, J. Am. Soc. Nephrol. 11 (2000) 1667–1673. [62] M.P. Cohen, E. Hud, J. Immunol. Meth. 117 (1989) 121–129. [63] J.M. Chamouard, J. Barre, S. Urien, et al., Biochem. Pharmacol. 34 (1985) 1695–1700. [64] A.M. Martinus, P. van Boekel, J.P. van den Bragh, et al., Biochim. Biophys. Acta 1120 (1993) 201–204. [65] R. Ajiboye, J.J. Harding, Exp. Eye Res. 49 (1989) 31–41. [66] M. Compton, D.C. Rixon, J.J. Harding, Exp. Eye Res. 40 (1985) 297–311. [67] R. Huby, J.J. Haring, Exp. Eye Res. 47 (1988) 53–59. [68] I. Jackle-Meyer, B. Szukics, E. Neubauer, et al., Eur. J. Clin. Chem. Clin. Biochem. 33 (1995) 211–219. [69] S. Kado, A. Aoki, S. Wada, et al., Diab. Res. Clin. Practice 31 (1996) 103–108. [70] M. Yagame, D. Suzuki, E. Jinke, et al., J. Lab. Clin. Anal. 11 (1997) 110–116. [71] N. Kotajima, T. Kimura, T. Kanada, et al., J. Diab. Complications 14 (2000) 13–17. [72] Y. Hayashi, H. Makino, Z. Ota, Diab. Med. 9 (1992) 366–370. [73] M.P. Cohen, G.T. Lautenslager, C.W. Shearman, Diabetes Care 24 (2001) 914–918. [74] M.P. Cohen, C.W. Shearman, G.T. Lautenslager, Diabetes Care 24 (2001) 1324–1327. [75] P. Fioretto, M.W. Steffes, M. Mauer, Diabetes 43 (1994) 1358– 1364. [76] S.M. Mauer, M.W. Steffes, N.E. Ellis, et al., J. Clin. Invest. 4 (1984) 1143–1155. [77] M.W. Steffes, R.W. Bilous, D.E.R. Sutherland, S.M. Mauer, Diabetes 38 (1999) 1077–1081. [78] M.P. Cohen, G.T. Lautenslager, C.W. Shearman, Metabolism 50 (2001) 1435–1440. [79] R. Osterby, H.H. Parving, E. Hommel, et al., Diabetes 39 (1990) 1057–1063. [80] F.N. Ziyadeh, K. Sharma, M. Ericksen, et al., Clin. Invest. 93 (1994) 536–542. [81] K. Sharma, F. Ziyadeh, Am. J. Physiol. 267 (1994) F1094– F1101. [82] M.P. Cohen, K. Sharma, J. Guo, et al., Exp. Nephrol. 6 (1998) 226–233. [83] T. Yamamoto, T. Nakamura, N.A. Nobel, et al., Proc. Natl. Acad. Sci. USA 90 (1993) 1814–1818. [84] F.N. Ziyadeh, B.B. Hoffman, C.E. Han, et al., Proc. Natl. Acad. Sci. USA 97 (2000) 8015–8020.
Minireview
Intervention strategies to prevent pathogenetic effects of glycated albumin Margo P. Cohen* Institute for Metabolic Research, University City Science Center, 3508 Market Street, Suite 420, Philadelphia, PA 19104, USA Received 25 January 2003, and in revised form 13 July 2003
Abstract Modification of proteins by nonenzymatic glycation is one of the underlying factors contributory to the development of complications of diabetes. In general, the nature of this structural modification falls into two broad categories: nonenzymatic glycation per se, which refers to the attachment of free carbohydrate to proteins in the Amadori construct, and Advanced Glycation Endproducts (AGE), which refers to a heterogeneous group of carbohydrate-modified products generated from the Amadori adduct by oxidation, polymerization, and other spontaneous reactions. This review will focus on the role of nonenzymatically glycated proteins, and in particular glycated serum albumin, in the pathogenesis of diabetic complications, and on pharmacologic approaches to mitigate their deleterious effects. Potential intervention strategies to lessen the influence of AGE-modified proteins, as well as of other contributory abnormalities, are discussed elsewhere in this volume. 2003 Elsevier Inc. All rights reserved.
The pathogenesis of the long-term complications of diabetes, which encompass microvascular, macrovascular, and neuropathic diseases, likely derives from diverse perturbations in cell biology superimposed on a polygenic predisposition and acting in concert with hyperglycemia, the defining metabolic abnormality of diabetes. Among the identified contributory processes are hemodynamic and oxidative stress, activation of various enzymatic pathways involved with glucose metabolism, stimulation of growth factor systems, activation of cell signaling pathways, modulation of responsivity to hormonal, growth factor, and cytokine influences, and acceleration of the nonenzymatic glycation of circulating and tissue proteins (Table 1). Understanding the role of and mechanisms by which these elements participate in the development of diabetic complications has advanced substantially in recent years. This review will focus on evidence linking increased nonenzymatic glycation of serum proteins to the pathophysiology of diabetic vasculopathies and potential avenues of therapeutic intervention to countermand the consequent deleterious effects. * Fax: 1-215-222-5325. E-mail address: [email protected].
0003-9861/$ - see front matter 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2003.08.012
Nonenzymatic glycation Nonenzymatic glycation is a condensation reaction between carbohydrate and free amino groups at the NH2 -terminus or susceptible e-amino groups of lysine residues of proteins. The reaction is initiated with attachment of the aldehyde function of acyclic glucose to the protein amino group via nucleophilic addition, forming an aldimine, also known as a SchiffÕs base. This intermediate product subsequently undergoes an Amadori rearrangement to form a 1-amino-deoxyfructose derivative in stable ketoamine linkage (Fig. 1), which in turn can cyclize to a ring structure [1–5]. The amount of the labile SchiffÕs base intermediate increases rapidly within a few hours after incubation of protein with glucose and reaches equilibrium [3], during which time the formation of the stable ketoamine is negligible. The rate of Amadori rearrangement has been calculated to be about 1/60th of the rate of dissociation of the aldimine back to glucose and protein, indicating that the rearrangement step is rate limiting for the formation of stable product. The glycation reaction follows secondorder kinetics and glucose is a major determinant of the amount of Amadori product formed. Another main determinant is the time of exposure of the protein to
26
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
Table 1 Pathogenetic factors in diabetic complications • • • • • • •
Polygenic predisposition Hemodynamic stress Oxidative stress Enzymatic shunting in glucose metabolism (polyol pathway; GFAT) Signal pathway activation (MAPK cascade; JAK/STAT) Stimulation of growth factor/cytokine systems (TGF-b; Angiotensin II: Thromboxane; Endothelins; IGF-I; PDGF: VEGF: Connective Tissue Growth Factor) Nonenzymatic glycation
Fig. 1. Condensation reaction between glucose and protein epsilon amino groups.
increased glucose concentration since glucoadducts will continue to form as a function of time until equilibrium is reached. In vivo, these two influences (glucose concentration and time) translate to degree and duration of hyperglycemia, and result in the slow and irreversible process of glycation during the life of the protein. Glycated albumin accounts for about 80% of the circulating glycated protein and albumin modified by Amadori glucose adducts is the predominant form in which glycated albumin exists in vivo. The principal residues subjected to glycation in vivo are lys-525, lys439, lys-281, and lys-199 [6,7]. Hyperglycemia attendant to the diabetic state increases nonenzymatically glycated albumin and the extent of increase reflects the time-averaged concentration of glucose to which the protein has been exposed during its residence time in the circulation. Like the measurement of the amount of glycated hemoglobin or HbA1c in red cells, determination of the serum concentration of glycated albumin assesses integrated glycemia over the relevant retrospective period, which is about 2–3 weeks in humans [4–6]. However, HbA1c serves only as a marker of glycemic status whereas glycated albumin has documented biologic effects, which have been causally linked to the pathogenesis of diabetic nephropathy and other complications. Moreover, as discussed below, these effects are
operative independent of hyperglycemia, indicating that they can continue to exert a pathogenetic influence in vivo during the half-life of the protein, even if normoglycemia is restored.
Biologic effects of glycated albumin The notion that exposure of the capillary beds to serum containing increased concentrations of glycated albumin due to diabetes could be deleterious to vascular physiology was suggested by early studies showing that transfusion into normal rats of Amadori-modified serum proteins in concentrations found in streptozotocin-diabetic rats induced hyperfiltration, a functional abnormality that has been implicated in the development of diabetic nephropathy [8]. Other earlier work in support of this hypothesis had indicated that glycated albumin is preferentially transported across the glomerular filtration barrier [9–11], that glomerular mesangial and epithelial cells exhibit an enhanced uptake of glycated albumin that is accompanied by an increase in cell hydrogen peroxide production [12,13], that nondiabetic mice injected with glycated plasma proteins develop glomerular basement membrane thickening [14], and that there is widespread connective tissue accumu-
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
27
Table 2 Comparability of biologic effects of glycated albumin and high glucose on glomerular cells in culture
Expression/Production of ECM proteins MAPK cascade Protein kinase C TGF-b system a b
Glycated albumina
Glucoseb
Stimulates Activates Activates Upregulates
Stimulates Activates Activates Upregulates
5.5 mM glucose media. P 25 mM glucose media.
lation of glycated albumin in diabetes [15–17]. More recent studies have uncovered a myriad of pathobiologic events induced by glycated albumin, which include increased gene expression of interleukin (IL)-8 and nitric oxide synthase and nitric oxide production in endothelial cells [18,19], activation of Protein Kinase C (PKC)1, the Mitogen Activated Protein Kinase (MAPK) cascade, and the Nuclear Factor (NF)-jB and increased gene expression of the inflammatory cytokines Monocyte Chemoattractant Peptide (MCP-1) and IL-6 in vascular smooth muscle cells [20,21], increased oxidative stress, phosphorylation of Extracellular Signal Regulated Kinase (ERK), and nuclear translocation of NFjB in monocyte–macrophages [22], and activation of signaling pathways and upregulation of growth factor systems in renal glomerular cells. Data from experiments with glomerular mesangial and endothelial cells in culture have established that albumin modified by Amadori glucose adducts induces significant alterations in glomerular cell biology that are highly reminiscent of abnormalities known to be associated with diabetic nephropathy and that resemble those induced by high glucose concentrations. Specifically, glycated albumin has been shown to: (a) stimulate the expression of a1 (IV) collagen and fibronectin, the predominant constituents of the expanded ECM that is seen in diabetes [23–28]; (b) activate PKC-b1 and ERK [29–31]; and (c) increase expression of mRNA encoding the fibrogenic Transforming Growth Factor (TGF)-b1 and its primary signaling receptor, the TGF-b type II receptor [27,31,32]. The modulations in cell biology induced by glycated albumin are seen with concentrations of glycated albumin representative of those found in clinical specimens and are analogous to but operate independent of those observed in response to high glucose concentration (Table 2), since they can be demonstrated in a physiologic (5.5 mM) glucose milieu. These in vitro data are consistent with the hypothesis that glycated albumin is a sufficient stimulus to set into motion a pathogenic program characterized by alterations in key cellular mediators that affect cell signaling pathways, 1 Abbreviations used: PKC, protein kinase C; MAPK, mitogen activated protein kinase; NF-jB, nuclear factor-jB; MCP-1, monocyte chemoattractant peptide; TGF-b1, transforming growth factor-b1; AGE, advanced glycation endoproducts.
which are important in regulating extracellular matrix (ECM) production. Together, the effects of glycated albumin and hyperglycemia may be additive, creating a self-reinforcing pathophysiologic cycle (Fig. 2). The hypothesis formulated from in vitro findings that glycated albumin is an independent and potent trigger of molecular mediators contributory to complications of diabetes is supported by results of in vivo studies showing that neutralizing its biologic effects [33–37] or inhibiting its formation [38–40] can ameliorate structural, cell biology, and functional abnormalities in the kidney and retinal microvasculature, and can do so in the face of persistent hyperglycemia.
Clinical considerations The traditional approach to forestalling the onset or progression of complications of diabetes has focused on reducing systemic risk factors and optimizing control of blood glucose, the potential benefits of the latter being underscored by the results of the Diabetes Control and Complications Trial (DCCT) and the more recent United Kingdom Prospective Diabetes Study (UKPDS) [41–44]. Despite wide publicity of the findings of these trials, which established that improved glycemic control reduces the risk for diabetic complications, the invigoration of efforts for bettering the glycemic status in diabetic patients, and the introduction of new anti-diabetic agents such as the thiazolidenediones, glycemic control remains woefully inadequate in the majority of people with diabetes. Normoglycemia is difficult to achieve and may be associated with hypoglycemic episodes of unacceptable frequency or severity [45]. Recent studies showing that modulation of hemodynamic influences with ACE inhibitors or angiotensin receptor (AR) antagonists can slow the inexorable progression of diabetic nephropathy offer promise [46–49], but these measures are neither curative nor completely effective. Thus, there is an appreciation that diabetic patients may require treatments specifically targeted at the vascular complications of the disease [50] and that, if possible, such agents should be effective, regardless of the glycemic status. Evidence from animal studies, discussed below, indicates that reduction of the burden of glycated albumin constitutes a viable therapeutic strategy to this
28
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
Fig. 2. Dual-pronged self-reinforcing pathogenetic processes triggered by glycated albumin and glucose. GA, glycated albumin, ECM, extracellular matrix.
end. Available clinical information also is supportive of this approach. In human diabetes, the concentration of Amadori-modified glycated albumin is independently associated with diabetic nephropathy and vascular dysfunction [51], and localization of Amadori-modified albumin in glomeruli of patients with diabetic nephropathy corresponds with the severity of renal involvement [52].
Therapeutic approaches Strong evidence supporting the view that glycated albumin is a potential target of therapy in the treatment of diabetic complications derives from studies in the db/ db mouse, a genetic model of diabetes that develops renal lesions resembling those observed in human diabetic nephropathy. The db/db mouse is usually considered a model of type 2 diabetes, but the early hyperglycemia, transition to an insulin deficient state, and development of classic renal pathology and functional abnormalities render it relevant for the study of diabetic nephropathy occurring in both type 1 and type 2 diabetes [53–59]. Diabetic nephropathy is best described in patients with type 1 diabetes, but similar structure–function changes are known to occur in patients with type 2 diabetes [60,61]. The first line of evidence comes from a series of studies examining the effects of treatment of db/db mice with a murine monoclonal antibody that specifically recognizes Amadori glucose adducts in rodent and human glycated albumin [62]. Chronic administration of the antibody was shown to reduce proteinuria and attenuate mesangial expansion, assessed by visual inspection, in diabetic db/db mice [33]. The reno-protective response to antibody treatment was interpreted to reflect a reduction in circulating biologically active glycated epitopes, since it was accompanied by a significant
decrease in the plasma glycated albumin concentration, ascribed to an enhanced clearance through the reticuloendothelial system. Subsequent experiments with rigorous glomerular morphometric analysis showed that treatment with the monoclonal antibody significantly reduced the accumulation of glomerular mesangial matrix, and that this reduction was accompanied by a decrease in the renal cortical overexpression of the extracellular matrix proteins fibronectin and type IV collagen, assessed by Northern blot [34]. Antibody treatment also prevented the thickening of the retinal microvessel basement membrane observed in control diabetic mice [37]. A similar program of chronic treatment with antibodies unreactive with glycated albumin was unable to duplicate these beneficial effects, corroborating the interpretation that neutralizing the biologic effects of excess Amadori-modified albumin in the circulation has a salutary influence on the development of renal and retinal pathology in diabetes. A second line of evidence comes from studies using small molecules that prevent the nonenzymatic glycation of albumin by binding to the protein at or near sites susceptible to glycation in vivo and rendering the involved e-amino groups inaccessible for condensation with glucose. The impetus to design such molecules was in part prompted by reports indicating that various antiinflammatory drugs could impede the in vitro attachment of free carbohydrate to protein [63–67]. New compounds have been synthesized that bind to wellcharacterized sites in serum albumin and that have potent anti-glycation properties and negligible cyclooxygenase inhibitory activity, and two of these have been shown to arrest the development of diabetic nephropathy in the db/db mouse [39,40]. After oral administration each of the compounds significantly lowered plasma concentrations of glycated albumin, confirming their ability to inhibit the condensation of glucose with glycatable sites in serum albumin in vivo.
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
They did not affect blood glucose concentrations, which would be an anticipated finding since the compounds do not react with free glucose and do not influence metabolic status. Despite persistent elevation of blood glucose concentrations, the drug-induced reductions in glycated albumin were accompanied by an amelioration of all of the functional and structural abnormalities characteristic of diabetic nephropathy, including significant decreases in the elevated urine protein excretion, the renal overexpression of ECM proteins, the mesangial matrix expansion, the falling creatinine clearance, and the rising serum creatinine. The in vivo inhibition of albumin glycation also lowered the elevated urinary excretion of type IV collagen, which reflects renal overproduction of this ECM protein and the associated glomerular matrix expansion that leads to glomerular occlusion and loss of filtration function in both human [68–77] and rodent [39,78,79] diabetes. Most notable is the finding that the treatment protocol significantly decreased the marked diabetes-associated glomerular overexpression of TGF-b1, assessed by in situ hybridization, thus closing the extrapolative loop between the in vitro observations that glycated albumin up-regulates the TGF-b system in glomerular mesangial and endothelial cells and the in vivo situation. The attenuation of the overexpression of glomerular TGF-b1 in conjunction with normalization of serum glycated albumin supports the hypothesis that increased glomerular TGFb1 expression in diabetes derives at least in part from the elevated circulating concentrations of glycated albumin associated with this disease. The data are consistent with the interpretation that the salutary effects on glomerulosclerosis (and, hence, compromised filtration function) associated with normalizing glycated albumin concentrations were mediated by a reduction in glomerular overexpression of TGF-b1, a potent fibrogenic cytokine that stimulates synthesis of ECM and inhibits its degradation and that has been strongly linked to diabetic glomerulosclerosis [80–84].
Conclusions and perspectives The studies from our laboratory and others that are recapitulated in this review have revealed that exposure of vascular beds to increased glycated serum proteins is an important contributor to the pathogenesis of diabetic complications. The evidence provided indicates that glycated albumin fulfills, to a large extent, KochÕs criteria for an etiologic agent in disease processes and that, once formed, its acquired biologic effects do not depend on a hyperglycemic milieu to be operative. Although hyperglycemia is the driving force for accelerated nonenzymatic glycation of proteins, it is clear from these studies that increased albumin glycation in diabetes can be therapeutically prevented in the presence
29
of hyperglycemia, and that therapeutic intervention to reduce the level of glycated albumin can arrest the evolution of diabetic nephropathy independent of antihyperglycemic therapy. The strategy of blocking the formation or effects of glycated albumin holds promise as a valuable therapeutic adjunct for the prevention and treatment of complications in human diabetes.
Acknowledgments This work was supported in part by research grants from the National Institutes of Health (DK54143; DK54608; and EY11825).
References [1] R.J. Koenig, S.H. Blobstein, A. Cerami, J. Biol. Chem. 256 (1981) 5204–5208. [2] H.F. Bunn, K.H. Gabbay, P.M. Gallop, Science 200 (1978) 21–27. [3] P.H. Higgins, H.F. Bunn, J. Biol. Chem. 256 (1981) 5204–5208. [4] J.F. Day, C.G. Ingelbretsen, W.R. Ingelbretsen, et al., Diabetes 29 (1980) 524–527. [5] C.E. Guthrow, M.A. Morris, J.F. Day, et al., Proc. Natl. Acad. Sci. USA 76 (1979) 4258–4261. [6] R.L. Garlick, J.S. Mazer, J. Biol. Chem. 258 (1983) 6142–6146. [7] N. Iberg, R. Fluckiger, J. Biol. Chem. 261 (1986) 13542–13546. [8] M. Sabbatini, G. Sansone, E. Ucello, et al., Kidney Int. 42 (1992) 875–881. [9] S.K. Williams, J.J. Devenny, M.W. Bitensky, Proc. Natl. Acad. Sci. USA 78 (1981) 2392–2397. [10] S.K. Williams, N.J. Solenski, Microvasc. Res. 28 (1984) 311–321. [11] S.K. Williams, R.K. Siegal, Kidney Int. 28 (1985) 146–152. [12] M. Agarwal, B.S. Daniels, Clin. Res. 40 (1992) 179A. [13] B.S. Daniels, R. Brandt, M. Agarwal, J. Am. Soc. Nephrol. 4 (1993) 791A. [14] B.A. McVerry, A. Hopp, C. Fisher, et al., Lancet 2 (1980) 738– 740. [15] K. Miller, A.F. Michael, Diabetes 25 (1976) 701–708. [16] A. Michael, D.M. Brown, Diabetes 30 (1981) 843–846. [17] S.B. Kelly, J.E. Olerud, J.L. Witztum, et al., J. Invest. Dermatol. 93 (1989) 327–331. [18] A. Amore, P. Cirina, S. Mitola, et al., Lab Invest. 51 (1997) 27–35. [19] Z.M. Bain, V.M. Elner, N.W. Lukacs, et al., Curr. Eye Res. 17 (1998) 65–72. [20] Y. Hattori, H. Kakishita, K. Akimoto, M. Matsumura, K. Kasai, Biochem. Biophys. Res. Commun. 281 (2001) 891–896. [21] Y. Hattori, M. Suzuki, S. Hattori, K. Kasai, Hypertension 39 (2002) 22–29. [22] M. Cohen, E. Shea, S. Chen, C.W. Shearman, J. Lab. Med. 141 (2003) 242–249. [23] M.P. Cohen, F.N. Ziyadeh, Kidney Int. 45 (1994) 475–484. [24] F.N. Ziyadeh, M.P. Cohen, Mol. Cell. Biochem. 125 (1993) 19–25. [25] M.P. Cohen, E. Hud, V.Y. Wu, F.N. Ziyadeh, Mol. Cell. Biochem. 151 (1995) 61–67. [26] M.P. Cohen, F.N. Ziyadeh, J. Am. Soc. Nephrol. 7 (1996) 183–190. [27] F.N. Ziyadeh, D.C. Han, J.A. Cohen, et al., Kidney Int. 53 (1998) 631–638. [28] M.P. Cohen, V.-Y. Wu, J.A. Cohen, Biochem. Biophys. Res. Commun. 139 (1997) 91–94.
30
M.P. Cohen / Archives of Biochemistry and Biophysics 419 (2003) 25–30
[29] M. Cohen, F.N. Ziyadeh, G.T. Lautenslager, et al., Am. J. Physiol. 276 (1999) F684–F690. [30] M.P. Cohen, E.A. Shea, C.W. Shearman, Biochem. Biophys. Res. Commun. 283 (2001) 641–643. [31] S. Chen, M.P. Cohen, G.T. Lautenslager, et al., Kidney Int. 59 (2001) 673–681. [32] S. Chen, M.P. Cohen, F.N. Ziyadeh, Kidney Int. 58 (2000) 40–44. [33] M.P. Cohen, E. Hud, V.Y. Wu, Kidney Int. 45 (1994) 1673–1979. [34] M.P. Cohen, K. Sharma, K. Jin, et al., J. Clin. Invest. 95 (1995) 2338–2345. [35] M.P. Cohen, R.S. Clements, J.A. Cohen, C.W. Shearman, Diabetologia 39 (1996) 270–274. [36] M.P. Cohen, R.S. Clements, J.A. Cohen, C.W. Shearman, Biochem. Biophys. Res. Commun. 218 (1996) 72–75. [37] R.S. Clements, W.G. Robison, M.P. Cohen, J. Diabetic. Complic. 12 (1998) 28–33. [38] M.P. Cohen, N. Mason, E. Hud, R.S. Clements, Exper. Nephrol. 8 (2000) 135–143. [39] M.P. Cohen, E. Hud, E. Shea, C.W. Shearman, Metabolism 51 (2002) 901–905. [40] M.P. Cohen, F.N. Ziyadeh, S.W. Hong, C.W. Shearman, et al., Kidney Int. 61 (2002) 2025–2032. [41] The DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin dependent diabetes mellitus, N. Engl. J. Med. 329 (1993) 977–986. [42] The DCCT Research Group: Effect of intensive therapy on the development and progression of diabetic nephropathy in the Diabetes Control and Complications Trial, Kidney Int. 47 (1995) 1703–1720. [43] UK Prospective Diabetes Study Group: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes, Lancet 352 (1998) 837–853. [44] UK Prospective Diabetes Study Group: Effect of intensive bloodglucose control with metformin on complications in overweight patients with type 2 diabetes, Lancet 352 (1998) 854–865. [45] The DCCT Research Group: Epidemiology of severe hypoglycemia in the diabetes Control and Complications Trial, Am. J. Med. 90 (1991) 450–459. [46] E.J. Lewis, L.G. Hunsicker, et al., N. Engl. J. Med. 329 (1993) 1456–1462. [47] E.J. Lewis, L.G. Hunsicker, et al., N. Engl. J. Med. 345 (2001) 851–860. [48] B.M. Brenner, M.E. Cooper, et al., N. Engl. J. Med. 354 (2001) 861–869. [49] T.H. Hostetter, N. Engl. J. Med. 354 (2001) 910–911. [50] E. Duh, L.P. Aiello, Diabetes 48 (1999) 1899–1906. [51] C.G. Schalkwijk, N. Ligtvoet, H. Twaalfhoven, et al., Diabetes 48 (1999) 2446–2453. [52] H. Sakai, K. Jinde, D. Suzuki, et al., Nephrol. Dial. Transplant. 11 (Suppl. 5) (1996) 66–71. [53] K. Hummel, M.M. Dica, D.L. Coleman, Science 153 (1966) 1127–1128.
[54] D.L. Coleman, K.P. Hummel, Diabetologia 1 (1974) 607–610. [55] W.L. Chick, A. Like, Diabetologia 6 (1970) 243–251. [56] A.A. Like, R.L. Levine, P.L. Poffenbarger, W.L. Chick, Am. J. Pathol. 66 (1972) 193–204. [57] S.M. Lee, A. Graham, Exp. Mol. Pathol. 33 (1980) 323–332. [58] S.M. Lee, R. Bressler, Diabetes 30 (1981) 106–111. [59] M.P. Cohen, R.S. Clements, E. Hud, et al., Exper. Nephrol. 4 (1996) 166–171. [60] A. Ueno, S. Kawashima, S. Nihsi, et al., Kidney Int. 54 (1997) S191–S194. [61] K.E. White, R.W. Bilous, J. Am. Soc. Nephrol. 11 (2000) 1667–1673. [62] M.P. Cohen, E. Hud, J. Immunol. Meth. 117 (1989) 121–129. [63] J.M. Chamouard, J. Barre, S. Urien, et al., Biochem. Pharmacol. 34 (1985) 1695–1700. [64] A.M. Martinus, P. van Boekel, J.P. van den Bragh, et al., Biochim. Biophys. Acta 1120 (1993) 201–204. [65] R. Ajiboye, J.J. Harding, Exp. Eye Res. 49 (1989) 31–41. [66] M. Compton, D.C. Rixon, J.J. Harding, Exp. Eye Res. 40 (1985) 297–311. [67] R. Huby, J.J. Haring, Exp. Eye Res. 47 (1988) 53–59. [68] I. Jackle-Meyer, B. Szukics, E. Neubauer, et al., Eur. J. Clin. Chem. Clin. Biochem. 33 (1995) 211–219. [69] S. Kado, A. Aoki, S. Wada, et al., Diab. Res. Clin. Practice 31 (1996) 103–108. [70] M. Yagame, D. Suzuki, E. Jinke, et al., J. Lab. Clin. Anal. 11 (1997) 110–116. [71] N. Kotajima, T. Kimura, T. Kanada, et al., J. Diab. Complications 14 (2000) 13–17. [72] Y. Hayashi, H. Makino, Z. Ota, Diab. Med. 9 (1992) 366–370. [73] M.P. Cohen, G.T. Lautenslager, C.W. Shearman, Diabetes Care 24 (2001) 914–918. [74] M.P. Cohen, C.W. Shearman, G.T. Lautenslager, Diabetes Care 24 (2001) 1324–1327. [75] P. Fioretto, M.W. Steffes, M. Mauer, Diabetes 43 (1994) 1358– 1364. [76] S.M. Mauer, M.W. Steffes, N.E. Ellis, et al., J. Clin. Invest. 4 (1984) 1143–1155. [77] M.W. Steffes, R.W. Bilous, D.E.R. Sutherland, S.M. Mauer, Diabetes 38 (1999) 1077–1081. [78] M.P. Cohen, G.T. Lautenslager, C.W. Shearman, Metabolism 50 (2001) 1435–1440. [79] R. Osterby, H.H. Parving, E. Hommel, et al., Diabetes 39 (1990) 1057–1063. [80] F.N. Ziyadeh, K. Sharma, M. Ericksen, et al., Clin. Invest. 93 (1994) 536–542. [81] K. Sharma, F. Ziyadeh, Am. J. Physiol. 267 (1994) F1094– F1101. [82] M.P. Cohen, K. Sharma, J. Guo, et al., Exp. Nephrol. 6 (1998) 226–233. [83] T. Yamamoto, T. Nakamura, N.A. Nobel, et al., Proc. Natl. Acad. Sci. USA 90 (1993) 1814–1818. [84] F.N. Ziyadeh, B.B. Hoffman, C.E. Han, et al., Proc. Natl. Acad. Sci. USA 97 (2000) 8015–8020.