Albumin modified by Amadori-glucose adducts induces coordinate increases in the expression of extracellular matrix proteins, transforming growth factor (TGF)-β1, and the TGF-β type II receptor in glomerular mesangial cells. Because activation of protein kinase C (PKC) accompanies the increased mesangial cell expression of matrix proteins and TGF-β1 induced by high ambient glucose, we postulated that glycated albumin (GA) modulates PKC activity and that PKC participates in mediating the GA-induced stimulation of matrix production. To test this hypothesis, we examined the effects of PKC inhibitors on collagen type IV production by mouse or rat mesangial cells incubated with GA, and the influence of GA on PKC activity in these cells. Increased collagen type IV production evoked by GA in 5.5 and 25 mM glucose in mouse mesangial cells was prevented by both general (GF-109203X) and β-specific (LY-379196) PKC inhibitors. Total PKC activity, measured by phosphorylation of a PKC-specific substrate, increased with time after exposure of rat mesangial cells to GA compared with the nonglycated, glucose-free counterpart. GA caused an increase in PKC-β1 membrane-bound fraction and in total PKC activity in media containing physiological (5.5 mM) glucose concentrations in rat mesangial cells, confirming that the glucose-modified protein, and not a “hyperglycemic” milieu, was responsible. The findings indicate that Amadori-modified albumin stimulates mesangial cell PKC activity, and that activation of the PKC-β isoform is linked to the stimulation of collagen type IV production.
- protein kinase C
- diabetic nephropathy
we have been investigating the pathobiology of diabetic nephropathy, with particular emphasis on the role of nonenzymatically glycated albumin (GA) and its molecular mediators in the accumulation of extracellular matrix, which is characteristic of this complication of diabetes. GA is formed from a condensation reaction between glucose and reactive protein amino groups, yielding stable Amadori-glucose adducts in a deoxyfructosyllysine construct (5). This modification confers biological properties to the glycated protein that are not possessed by its nonglycated counterpart and that can modulate glomerular cell function. Glomerular mesangial and endothelial cells incubated in concentrations of GA that are found in clinical specimens manifest increased gene expression of the extracellular matrix proteins collagen type IV and fibronectin (9, 12, 13, 38). Further, GA induces increased mesangial cell expression and bioactivation of the fibrogenic transforming growth factor (TGF)-β1 and its primary signaling receptor, the TGF-β type II receptor, thus linking activation of the TGF-β system to the glycated protein-induced stimulation of extracellular matrix production (41). Additional evidence supporting a role for GA in the genesis of matrix overproduction in diabetes derives from experiments in which diabeticdb/db mutant mice were treated with monoclonal antibodies (A717) that specifically neutralize excess plasma concentrations of GA. This protocol prevented renal cortical overpression of mRNAs encoding α1(IV) collagen and fibronectin, reduced mesangial matrix expansion, and improved changes reflecting compromised renal function (6, 8, 10). These salutary effects were observed despite marked and persistent hyperglycemia.
Recent studies have implicated the activation of protein kinase C (PKC) in the abnormalities in glomerular cell function associated with diabetes (16). PKC is a ubiquitous family of related intracellular serine-threonine kinases that are involved in signal transduction pathways used by cells to respond to various extracellular stimuli, and in the regulation of diverse cellular functions, such as contractility, proliferation, and hemodynamic control (32). Evidence suggesting a role for PKC in the pathogenesis of diabetic nephropathy includes the observations that mesangial cells cultured in high media glucose concentration exhibit increased PKC activity and a cytosol-to-membrane shift of various classical PKC isoforms that is associated with activation (2, 29), and that glomeruli from diabetic rodents exhibit elevated PKC activity and membrane translocation (3, 16, 26, 30). Increased activity of PKC in mesangial cells is associated with increased expression of TGF-β1 (15, 23) and of the extracellular matrix proteins fibronectin, laminin, and collagen type IV (15, 17, 30,35, 40). Inhibition of PKC-β isoforms in streptozotocin-diabetic rats has been reported to ameliorate glomerular hyperfiltration, increased urine albumin excretion (26), and the stimulation of gene expression for TGF-β1 and matrix molecules in the kidney (30).
Because the alterations in glomerular cell biology induced by GA resemble those evoked by high media glucose concentration, we postulated that GA might also activate PKC isoforms in glomerular mesangial cells. To test this hypothesis, we examined the effects on PKC activity of Amadori-modified albumin, the principal form in which GA exists in vivo (5, 13). We report that PKC activity increases when mesangial cells are incubated in concentrations of Amadori-modified GA that are found in clinical specimens, even when the glucose concentration in the culture media is normal (5.5 mM), and that PKC signaling, particularly through the PKC-β isoform, participates in the increased collagen type IV production by these cells on exposure to GA.
MATERIALS AND METHODS
Mesangial cells in culture were derived from glomeruli isolated with a graded sieving technique and plated to explant. Mesangial cells were selected and subcultured according to previously described criteria (13), including morphology, ability to grow in medium containingd-valine instead ofl-valine, and the presence of cytoplasmic filaments (desmin and vimentin), angiotensin II binding capacity, and contractile response to angiotensin II. Phenotypically stable nontransformed rat (RMC) or SV40-transformed murine mesangial cells (MMC) were grown in DMEM containing 10 mM glucose and 10% FCS. MMC were employed for studies of collagen synthesis because their parallelism with RMC, with respect to growth and response to elevated glucose and GA media manipulations, has been documented (13, 38,41). RMC were employed for studies of PKC activity and immunoreactivity because preliminary experiments indicated that these cells yielded reproducible and reliable measurements with the PKC assay and immunoblotting techniques. To initiate experiments for collagen synthesis, we seeded 2 × 105cells into 48-well microtiter plates, allowed them to attach, rested them for 24 h in serum-free media containing 10 mM glucose, and then grew them for 48 h in fresh media under the experimental conditions described below. For assay of PKC activity, 2 × 105 mesangial cells were seeded into 24-well microtiter plates. After attachment, the cells were grown for 24 h in DMEM/10 mM glucose/2% FCS. The media were then changed to 1% FCS containing 5.5 or 25 mM glucose, and the cells were grown for 6–48 h under the described experimental conditions.
Experimental culture conditions.
The experimental conditions were introduced on the addition of fresh media containing 5.5 mM or 25 mM glucose, without or with the described supplements in the indicated concentrations. Media supplements consisted of purified glycated or nonglycated albumin (250–600 μg/ml) and the general PKC inhibitor GF-109203X (500 nM) or the selective PKC-β inhibitor LY-379196 (100 nM) (a gift to F.N. Ziyadeh from Eli Lilly). LY-379196 was added at a 100-nM concentration, because β-specific inhibitory activity may be lost if this concentration is exceeded (ED50, ∼600 nM for nonspecific PKC inhibition; ED50, ∼30 nM for selective PKC-β inhibition) (28). For studies of collagen production, media also contained 50 μg each ofl-ascorbic acid and β-aminopropionitrile (43). In the presence of β-aminopropionitrile, very little collagen type IV synthesized by mesangial cells in culture is cell associated, and ∼80% is recovered in the media (22). The concentrations of GA used in these studies have been shown to stimulate glomerular endothelial and mesangial cell expression of collagen type IV, fibronectin, and TGF-β1 (9, 12, 39, 41), and represent those found in clinical specimens. In nondiabetic individuals, ∼1% of serum albumin is in the glycated form, which is equivalent to concentrations of 300–400 μg/ml of GA. The concentration of GA is increased one-and-a-half- to threefold in diabetic subjects, according to recent glycemic status (5, 7).
Glycated protein preparation.
GA was prepared from human albumin that was purified by chromatography on Affi-gel Blue and DEAE-Sepharose, and incubated for 5 days at 25°C in buffered saline containing 500 mg/dl (27.8 mM) glucose. After dialysis to remove free glucose, the glycated species were separated from nonglycated albumin by affinity chromatography on phenylboronate agarose (PBA), which binds Amadori adducts and not advanced glycation end products (AGE). This protocol has been shown to yield GA containing ∼1 mol glucose/mol albumin and in which glycated moieties are represented as deoxyfructosyllysine residues (9, 39). The PBA pass through, used as source material for nonglycated albumin in these experiments, contained <0.05 mol glucose/mol albumin. The purified glycated and nonglycated albumin preparations migrated on SDS-PAGE as homogeneous bands of ∼66 kDa, and had distinct mobilities on agarose gel electrophoresis, wherein the glycated protein exhibited greater electronegativity consequent to glycation of lysine amino groups.
Collagen type IV measurement.
Media were collected at the end of the experimental period and the cells were harvested for counting in a cell chamber. We measured collagen type IV by competitive ELISA (13, 39) using collagen type IV from an Engelbreth-Holm-Swarm (EHS) tumor as standard, rabbit anti-mouse collagen type IV as primary antibody (both from Collaborative Research), and horseradish peroxidase-conjugated goat anti-rabbit IgG (Bio-Rad) for development. The assay is sensitive to 5 ng/well.
A modification of the method described by Thomas et. al (36) was used for analysis of immunoreactive PKC-β in particulate (membrane-associated) and cytosolic fractions prepared from RMC. After washing the cells three times with ice-cold PBS, we added 600 μl of ice-cold homogenization buffer [10 mM Tris ⋅ HCl, pH 7.5, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM EGTA, and 25 μg/ml leupeptin] to each sample. Cells were lifted with a cell scraper and transferred, in the buffer, to a homogenizer. The homogenized preparations were centrifuged at 2,000 rpm for 5 min at 4°C. The supernatants were separated from undisrupted cellular elements remaining in the pellet and centrifuged at 100,000g at 4°C for 1 h. The supernatants from this centrifugation were retained as the cytosolic fractions, and the pellet was suspended in the above homogenization buffer, made 1% in Triton X-100, and shaken for 30 min at 4°C to solubilize particulate proteins. This material was then centrifuged at 10,000 rpm for 1 h at 4°C, and the supernatants were retained as the particulate fractions. Protein was measured with the Bio-Rad (Hercules, CA) protein assay.
Immunoblotting was performed according to described techniques (1), with some modifications. Samples were electrophoresed on 8% SDS-PAGE and transferred to nitrocellulose membranes. Prestained molecular weight markers were electrophoresed in parallel. Transfers were blocked in a solution of 5% nonfat milk, 5 mM Tris, pH 7.5, 200 mM NaCl, and 0.1% Triton X-100, and then probed with primary antibody (rabbit anti-PKC-β1; GIBCO). Incubation with primary antibody was conducted overnight in 1% milk, 50 mM Tris, pH 8.0, 200 mM NaCl, and 0.1% Tween. After washing, transfers were developed with horseradish peroxidase-conjugated anti-rabbit IgG and the enhanced chemiluminescent detection system (Amersham), followed by exposure to X-Omat film (Eastman Kodak, Rochester, NY). Equal loading and transfer of samples was assessed by staining with Ponceau Red. Authentic PKC-β1 peptide was electrophoresed, transferred, and immunoblotted to serve as the control.
PKC activity was measured with the PepTag assay system (Promega), in which the change in charge of a fluorescent-tagged, PKC-specific substrate (PLSRTLSVAAK), which occurs with phosphorylation, is detected on separation with agarose gel electrophoresis at a neutral pH. At the end of the incubation period, the cells were washed with Hanks’ balanced salt solution containing 1 mM PMSF and 1 μM leupeptin, placed in 100 μl of the same buffer, and then freeze thawed for lysis. Aliquots were taken for the assay, which was performed according to the manufacturer’s instructions. Photographs of the gels were scanned into a densitometer program for quantitation (Scion Image; National Institutes of Health). Each assay included a positive control for PKC activity, wherein PKC supplied by the manufacturer was simultaneously subjected to the assay and electrophoretic procedure. A negative control also was run with each assay.
The purpose of these experiments was to determine whether Amadori-modified GA modulates PKC activity in glomerular mesangial cells, and to probe whether PKC signaling participates in the increased collagen production by these cells on exposure to GA. Because high media glucose concentration has been reported to activate PKC and to augment matrix production in glomerular cells, incubations were conducted in 5.5 mM glucose concentration to ensure that observed responses did not accrue from an independent effect of elevated media glucose. Where indicated, incubations in media containing 25 mM glucose were performed in parallel for comparative purposes and to document methodological integrity of the PKC assay by demonstration of high glucose-stimulated increases. Parallel incubations with nonglycated albumin, in the same concentrations as the glycated protein, in 5.5 or 25 mM glucose, served as controls. Under the conditions employed, cell counts were not significantly different in incubations containing glycated vs. nonglycated protein, as has been previously reported (12).
The concentration of collagen type IV in media from mouse mesangial cells, incubated with GA in 5.5 mM glucose, was significantly increased compared with that in the media of cells cultured in 5.5 mM glucose in the presence of nonglycated albumin (Fig.1). Relative to cells grown in 5.5 mM glucose with the nonglycated protein, these increments were dose related at 190% and 230% of control at the concentrations of GA studied. High media glucose concentration (25 mM) also stimulated collagen type IV production. Collagen type IV concentration in media from cells grown in 25 mM glucose was 170% of that in cells incubated with 5.5 mM glucose. Collagen type IV concentrations were further increased when cells were cultured in 25 mM glucose in the presence of GA. Relative to cells cultured with GA in 5.5 mM glucose, this increment reached significance at the highest concentration of GA studied, consistent with an additive effect of high ambient glucose and GA on matrix protein synthesis (Fig. 1).
In the presence of the general PKC inhibitor GF-109203X, concentrations of collagen type IV in mouse mesangial cells cultured in 5.5 mM glucose and nonglycated albumin did not differ from those in media from cells cultured without the inhibitor under the same conditions. However, GF-109203X prevented the increase in collagen type IV production induced by 25 mM glucose or by GA under both low and high media glucose conditions (Fig. 2). Similarly, the selective PKC-β inhibitor LY-379196 did not affect production of collagen type IV by cells grown in 5.5 mM glucose and nonglycated albumin, but prevented the GA-induced increases in collagen type IV production in cells incubated with either 5.5 mM or 25 mM glucose and reduced the collagen type IV production in cells incubated with 25 mM glucose in the presence of nonglycated albumin (Fig. 2).
The presence of PKC-β isoforms in rat mesangial cells was confirmed with immunoblotting experiments. Additionally, because classical members of the PKC family move to membranes on activation, and because a cytosol-to-membrane shift of PKC isoenzymes suggests activation of that isoenzyme, we examined PKC-β1 immunoactivity in the cytosolic and particulate fractions of mesangial cells to evaluate whether incubation with GA was associated with cytosol-to-membrane translocation. Immunoblots with anti-PKC-β1 antibodies after incubation of RMC (Fig. 3) or MMC (data not shown) confirmed immunoreactivity in both membranes and cytosolic fractions. Further, Fig. 3 shows that, under normoglycemic conditions, incubation with GA increased PKC-β1 immunoactivity in the membrane-bound fraction (with little discernible change in the cytosolic fraction), consistent with activation and, possibly, translocation of this isoenzyme.
Stimulation of PKC activity by GA was confirmed by measurement in the phosphorylation assay. Visual inspection of gels from experiments in which rat mesangial cells were incubated for 48 h in 5.5 mM glucose indicated that PKC activity was increased in cells cultured in media containing GA compared with activity in cells incubated with nonglycated albumin in the same concentrations (Fig.4). PKC activity in cells incubated for 48 h in 25 mM glucose was greater than that in cells cultured in 5.5 mM glucose, but it was difficult to appreciate visually whether GA had an additional effect in high glucose medium (Fig. 4). However, densitometric scanning of the gels confirmed that 48 h exposure to GA induced a dose-related increase in PKC activity under both low (5.5 mM) and high (25 mM) glucose conditions (Fig.5). PKC activity in cells incubated in 5.5 mM glucose with nonglycated albumin was not significantly different from that in 5.5 mM glucose alone, whereas the relative ratios (activity compared with 5.5 mM glucose alone, assigned an arbitrary value of 1.0) with 250 and 500 μg/ml of GA were 1.58 and 1.86, respectively. PKC activity in cells incubated for 48 h with 25 mM glucose was 169% of that in cells incubated for the same period in 5.5 mM glucose. In the presence of GA, the relative ratios (activity compared with 25 mM glucose alone, assigned an arbitrary value of 1.0) with 250 and 500 μg/ml of GA were 1.38 and 1.85, respectively. The relative ratios when activity was compared with 5.5 mM glucose, assigned a value of 1.0, were 2.33 and 3.13 for 250 and 500 μg/ml GA, respectively (Fig. 5).
Time-course studies in rat mesangial cells revealed that media containing either 25 mM glucose concentration or GA in 5.5 mM glucose concentration elicited steady increases in PKC activity over the 48-h period studied (Fig. 6). Relative ratios of PKC activity (compared with 5.5 mM glucose alone) in cells incubated for 6, 18, and 48 h in 5.5 mM glucose with 500 μg/ml of GA were 1.22, 1.51, and 1.86, respectively. The relative ratios of PKC activity in cells incubated for these periods in 25 mM glucose (compared with 5.5 mM glucose alone) were 1.13, 1.37, and 1.69, respectively. The steady increases in activity under high glucose conditions were consistent with a requirement for cellular de novo production of diacylglycerol in the mediation of glucose-induced effects on PKC activity. The slopes of the time-course lines of PKC activity under GA/5.5 mM glucose vs. 25 mM glucose (without GA) were almost identical (m = 0.013 and 0.014, respectively), in keeping with the foregoing results indicating that the increases in PKC activity stimulated by high glucose alone or by GA (in low glucose media) were comparable. Although the slope of the time-course line of PKC activity under conditions of GA/25 mM glucose was lower (m = 0.009), values at each time interval were higher than with 25 mM glucose alone (Fig. 6), consistent with an additive effect, as illustrated in the experiments depicted in Fig. 5.
The results of these experiments demonstrate that GA stimulates PKC activity in renal glomerular mesangial cells, and indicate that activation of selective PKC isoforms participates in the GA-induced stimulation of extracellular matrix protein production by these cells. This interpretation is based on the observations that PKC activity increases in rat mesangial cells incubated with glycated, but not nonglycated, albumin; that the glycated protein increases membrane-associated PKC-β1 immunoreactivity; and that inhibitors of PKC prevent the increases in collagen type IV production that occur on exposure of mouse mesangial cells in culture to GA. In this regard, both the general PKC inhibitor GF-109203X and the selective PKC-β inhibitor LY-379196 were effective in 5.5 mM glucose concentration, suggesting that PKC-β isoforms predominate in mediating the signal from a GA stimulus for matrix overproduction. It should be noted that LY-379196 does not differentiate between the two splice variants of PKC-β enzyme, though our studies in mesangial cells indicate that GA can selectively activate PKC-β1, one of the splice variants.
TGF-β1 is a key cytokine for matrix protein production (23, 34, 41,42). GA has been previously shown to induce coordinate increases in mRNAs encoding TGF-β1, the TGF-β type II receptor, and the extracellular matrix proteins fibronectin and collagen type IV (9, 13,39, 41). Given that PKC activation is believed to promote gene expression of TGF-β1 and extracellular matrix proteins, the findings of the current study suggest that GA-induced activation of PKC may be linked to these responses. This chain of events resembles that induced by high media glucose concentration, which increases mesangial cell PKC activity in association with increased gene expression of TGF-β1 and extracellular matrix proteins (2, 15, 17, 23, 29, 30, 34, 40). However, in the present experiments, increased PKC activity and collagen type IV production were observed in a low physiological glucose concentration (5.5 mM), and cannot be ascribed to an effect of high glucose concentration in the culture media.
The activation of PKC by high glucose is believed to derive from conversion of glycolytic intermediates into increased de novo production of diacylgycerol, a major endogenous activator of some PKC species (2, 14, 20). PKC activation triggers a series of phosphorylation events, including mitogen-activated protein kinases (19, 21, 23, 27) and increased expression ofc-fos andc-jun protooncogenes (31), which form a heterodimer, AP-1, that activates the TGF-β1 gene promoter (4, 23). The mechanism by which GA promotes PKC activation is speculative, but may relate to an interaction of the Amadori-modified protein with cell-associated ligand receptor systems (11). Knowing whether GA acts through diacylglycerol will require further study. In this regard, it is interesting to note that our studies have shown that GA induces sustained activation of PKC, which contrasts with the transient response effectuated by hormonal stimuli or by synthetic diglycerides, and is consistent with the persistent PKC membrane translocation that has been observed in tissues from diabetic animals (1, 25, 26, 37).
The characterization and cellular localization of PKC isoforms, and changes induced by high glucose or associated with diabetes, have not been completely delineated. Constitutive expression of several classical PKC isoforms in glomerular mesangial and epithelial cells has been reported, with some inconsistency regarding the presence of PKC-β1 and -β2 in mesangial cells (3, 18, 23, 24, 29, 30). Membrane-associated PKC-β2 has been reported to be increased (26) and decreased (3) in glomeruli from rats with acute streptozotocin diabetes, whereas membrane-associated PKC-β1 has been reported to be increased after 12 wk of streptozotocin diabetes (30). PKC-α has been more consistently identified in mesangial cells (18, 23, 29), and membrane associated PKC-α has been found to be increased in glomeruli of streptozotocin-diabetic rodents after 2, 4, and 12 wk of diabetes (3, 30). Our findings that GA increases PKC activity, and that PKC-β activation participates in GA-induced stimulation of matrix synthesis (as evidenced by prevention of this stimulation with both general and PKC-β-specific inhibitors) resemble observations with mesangial cells incubated with high media glucose concentration and in glomeruli from diabetic rats. However, it should be noted that, in the context of the above, the relative importance of the different PKC isoforms in normal and pathological states may differ in different glomerular cell types.
The findings reported herein are the first to demonstrate that albumin modified by Amadori glucose-adducts modulates PKC isoenzyme activity, and does so in a “normoglycemic” milieu and in concentrations of GA that are typically found in serum specimens of diabetic patients. It is likely that the results obtained in mouse and rat mesangial cells can be extrapolated to human mesangial cells, but direct experimental confirmation is required. These observations afford new insight into the PKC activation that has been found in tissues from diabetic animals, and into pathophysiological mechanisms contributory to extracellular matrix accumulation in diabetic renal disease. Either elevated glucose or GA can activate mesangial cell PKC, and either can induce TGF-β1 gene expression, which stimulates extracellular matrix protein production. Hyperglycemia is the driving force for increased albumin glycation, but the effects of GA on PKC activation and TGF-β1 expression do not require a hyperglycemic milieu to be operative. Thus the in vivo influence of GA, which has a circulating half-life of ∼2 wk, on PKC isoenzymes can continue after restoration of normoglycemia and can act independently of hyperglycemia. We have shown that therapy directed against increased GA beneficially influences nephropathology in a rodent model of genetic diabetes, despite marked and persistent hyperglycemia (6, 8, 10). We postulate that the salutary effects of this intervention strategy may derive, at least in part, from a reduction of PKC activity concomitant with a lowering of circulating GA concentrations. This approach affords the potential advantage of avoiding possible untoward effects of systemic PKC inhibition (33).
In summary, we have shown that albumin modified by Amadori glucose-adducts stimulates PKC activity and collagen type IV production in rodent mesangial cells in culture. These effects are observed in physiological glucose concentration. The findings link PKC signaling, the PKC-β isoform in particular, to the increased extracellular matrix production induced by elevated concentrations of GA in diabetes mellitus.
We thank Dr. D. K. Ways for the gift of LY-379196 (Lilly Research Laboratories). The technical assistance of Ryan Grinkewitz is gratefully acknowledged.
Address for reprint requests and other correspondence: F. N. Ziyadeh, Renal-Electrolyte & Hypertension Div., Univ. of Pennsylvania, 700 Clinical Research Bldg., 415 Curie Blvd., Philadelphia, PA 19104-6144 (E-mail:).
This work was supported by National Institutes of Health Grants DK-54608, DK-54143, DK-49455, DK-45191, DK-44513, and EY-11825.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 1999 the American Physiological Society